WO2020054591A1

WO2020054591A1 - Encoding device, decoding device, encoding method, and decoding method

Info

Publication number: WO2020054591A1
Application number: PCT/JP2019/035125
Authority: WO
Inventors: ジンヤリ; ルリンリャオ; チョンスンリム; ハンブンテオ; ハイウェイスン; 安倍　清史; 西　孝啓; 遠間　正真
Original assignee: パナソニックインテレクチュアルプロパティコーポレーションオブアメリカ
Priority date: 2018-09-14
Filing date: 2019-09-06
Publication date: 2020-03-19
Also published as: US11350111B2; US20230396776A1; US20210218970A1; JPWO2020054591A1; EP4054191A1; JP2022058478A; JP7004835B2; EP3852372A1; EP3852372A4; CN112840659B; JP2024036476A; BR112021001890A2; JP2023029589A; US11778203B2; KR20210052464A; MX2024004941A; MX2021002510A; JP7432022B2; US20220248033A1; JP7212181B2

Abstract

An encoding device (100) is provided with a circuit and a memory connected to the circuit. The circuit: derives a gradient sum absolute value that is the absolute value of the sum of a gradient value in a first range and a gradient value in a second range; derives, as a first parameter, the sum total of a plurality of gradient sum absolute values derived, respectively, for a plurality of relative pixel positions; derives a pixel difference value which is the difference between a pixel value in the first range and a pixel value in the second range; reverses or maintains the positive/negative sign of the pixel difference value depending on the positive/negative sign of a gradient sum, which is the sum of the gradient value in the first range and the gradient value in the second range; derives, as a second parameter, the sum total of a plurality of the pixel difference values which were respectively derived for the plurality of relative pixel positions and for which the positive/negative sign was respectively reversed or maintained; and uses the first parameter and the second parameter to generate a prediction image.

Description

Encoding device, decoding device, encoding method, and decoding method

The present disclosure relates to video coding, for example, to systems, components, and methods in video encoding and decoding.

Video coding technology is H.261 and MPEG-1, H.264 / AVC (Advanced Video Coding), MPEG-LA, H.264. H.265 / HEVC (High \ Efficiency \ Video \ Coding); 266 / VVC (Versatile Video Video Codec). With this advancement, there is a constant need to provide improvements and optimizations in video coding techniques to handle the ever-increasing amounts of digital video data in various applications.

Note that Non-Patent Document 1 relates to an example of a conventional standard relating to the above-described video coding technology.

Regarding the above-mentioned coding schemes, the coding efficiency is improved, the image quality is improved, the processing amount is reduced, the circuit scale is reduced, or elements or operations such as filters, blocks, sizes, motion vectors, reference pictures or reference blocks, etc. Proposal of a new method is desired for appropriate selection and the like.

The present disclosure can contribute, for example, to one or more of improvement of coding efficiency, improvement of image quality, reduction of processing amount, reduction of circuit size, improvement of processing speed, and appropriate selection of elements or operations. An arrangement or method is provided. Note that the present disclosure may include configurations or methods that can contribute to benefits other than the above.

For example, an encoding device according to an embodiment of the present disclosure includes a circuit, and a memory connected to the circuit, wherein the circuit operates in a first range including a first reference block of a current block, and A plurality of pixel positions that are commonly and relatively determined for both of a second range including a second reference block of the current block, and are a plurality of pixel positions in each of the first range and the second range. For each of the plurality of relative pixel positions, a gradient sum absolute value that is an absolute value of a sum of a gradient value of the relative pixel position in the first range and a gradient value of the relative pixel position in the second range. Is derived, and a total sum of a plurality of gradient sum absolute values derived respectively for the plurality of relative pixel positions is derived as a first parameter, and each of the plurality of relative pixel positions is Then, a pixel difference value that is a difference between the pixel value of the relative pixel position in the first range and the pixel value of the relative pixel position in the second range is derived, and the pixel difference value of the plurality of relative pixel positions is derived. In each case, the relative pixel position is derived by the sign of the gradient sum, which is the sum of the gradient value of the relative pixel position in the first range and the gradient value of the relative pixel position in the second range. Inverting or maintaining the sign of the obtained pixel difference value, derived for each of the plurality of relative pixel positions, and the sum of the plurality of pixel difference values in which the sign is inverted or maintained, respectively, as a second parameter. And generating a predicted image used for encoding the current block using the first parameter and the second parameter.

Some implementations of the embodiments in the present disclosure may improve encoding efficiency, simplify encoding / decoding processing, and increase encoding / decoding processing speed. , Suitable components / operations used for encoding and decoding, such as appropriate filters, block sizes, motion vectors, reference pictures, reference blocks, etc. may be efficiently selected.

さらなる Additional advantages and effects of one embodiment of the present disclosure will be apparent from the description and the drawings. Such advantages and / or effects may be obtained by means of some embodiments and features described in the specification and drawings, but not all are provided to achieve one or more advantages and / or effects. Need not be.

Note that these general or specific aspects may be realized by a system, a method, an integrated circuit, a computer program, a recording medium, or any combination thereof.

The configuration or method according to an aspect of the present disclosure includes, for example, improvement of encoding efficiency, improvement of image quality, reduction of processing amount, reduction of circuit scale, improvement of processing speed, and appropriate selection of elements or operations. Can contribute to one or more of them. Note that the configuration or method according to one embodiment of the present disclosure may contribute to benefits other than those described above.

FIG. 1 is a block diagram showing a functional configuration of the encoding device according to the embodiment. FIG. 2 is a flowchart illustrating an example of an overall encoding process performed by the encoding device. FIG. 3 is a conceptual diagram illustrating an example of block division. FIG. 4A is a conceptual diagram illustrating an example of the configuration of a slice. FIG. 4B is a conceptual diagram illustrating an example of a tile configuration. FIG. 5A is a table showing transformation basis functions corresponding to various transformation types. FIG. 5B is a conceptual diagram showing an example of SVT (Spatially Varying Transform). FIG. 6A is a conceptual diagram illustrating an example of the shape of a filter used in an ALF (adaptive loop filter). FIG. 6B is a conceptual diagram illustrating another example of the shape of the filter used in the ALF. FIG. 6C is a conceptual diagram showing another example of the shape of the filter used in the ALF. FIG. 7 is a block diagram illustrating an example of a detailed configuration of a loop filter unit that functions as a DBF (deblocking filter). FIG. 8 is a conceptual diagram showing an example of a deblocking filter having filter characteristics symmetric with respect to a block boundary. FIG. 9 is a conceptual diagram for explaining a block boundary where deblocking filter processing is performed. FIG. 10 is a conceptual diagram illustrating an example of the Bs value. FIG. 11 is a flowchart illustrating an example of a process performed by the prediction processing unit of the encoding device. FIG. 12 is a flowchart illustrating another example of the processing performed by the prediction processing unit of the encoding device. FIG. 13 is a flowchart illustrating another example of the processing performed by the prediction processing unit of the encoding device. FIG. 14 is a conceptual diagram illustrating an example of 67 intra prediction modes in intra prediction according to the embodiment. FIG. 15 is a flowchart illustrating an example of a basic processing flow of inter prediction. FIG. 16 is a flowchart illustrating an example of motion vector derivation. FIG. 17 is a flowchart illustrating another example of deriving a motion vector. FIG. 18 is a flowchart illustrating another example of deriving a motion vector. FIG. 19 is a flowchart illustrating an example of inter prediction in the normal inter mode. FIG. 20 is a flowchart illustrating an example of inter prediction in the merge mode. FIG. 21 is a conceptual diagram illustrating an example of a motion vector derivation process in the merge mode. FIG. 22 is a flowchart illustrating an example of FRUC (frame @ rate @ up @ version) processing. FIG. 23 is a conceptual diagram illustrating an example of pattern matching (bilateral matching) between two blocks along a motion trajectory. FIG. 24 is a conceptual diagram illustrating an example of pattern matching (template matching) between a template in a current picture and a block in a reference picture. FIG. 25A is a conceptual diagram illustrating an example of deriving a motion vector in sub-block units based on motion vectors of a plurality of adjacent blocks. FIG. 25B is a conceptual diagram for explaining an example of deriving a motion vector in subblock units in the affine mode having three control points. FIG. 26A is a conceptual diagram for explaining the affine merge mode. FIG. 26B is a conceptual diagram illustrating an affine merge mode having two control points. FIG. 26C is a conceptual diagram illustrating an affine merge mode having three control points. FIG. 27 is a flowchart illustrating an example of the affine merge mode process. FIG. 28A is a conceptual diagram for describing an affine inter mode having two control points. FIG. 28B is a conceptual diagram illustrating an affine inter mode having three control points. FIG. 29 is a flowchart illustrating an example of the affine inter mode processing. FIG. 30A is a conceptual diagram illustrating an affine inter mode in which a current block has three control points and an adjacent block has two control points. FIG. 30B is a conceptual diagram illustrating an affine inter mode in which a current block has two control points and an adjacent block has three control points. FIG. 31A is a flowchart showing a merge mode including DMVR (decoder \ motion \ vector \ refinement). FIG. 31B is a conceptual diagram illustrating an example of the DMVR process. FIG. 32 is a flowchart illustrating an example of generation of a predicted image. FIG. 33 is a flowchart illustrating another example of generation of a predicted image. FIG. 34 is a flowchart illustrating another example of generation of a predicted image. FIG. 35 is a flowchart for explaining an example of a predicted image correction process by an OBMC (overlapped \ block \ motion \ compensation) process. FIG. 36 is a conceptual diagram for describing an example of a predicted image correction process by the OBMC process. FIG. 37 is a conceptual diagram for describing generation of a predicted image of two triangles. FIG. 38 is a conceptual diagram for explaining a model assuming uniform linear motion. FIG. 39 is a conceptual diagram illustrating an example of a predicted image generation method using luminance correction processing by LIC (local illumination compensation) processing. FIG. 40 is a block diagram illustrating an implementation example of an encoding device. FIG. 41 is a block diagram showing a functional configuration of the decoding device according to the embodiment. FIG. 42 is a flowchart illustrating an example of the entire decoding process performed by the decoding device. FIG. 43 is a flowchart illustrating an example of a process performed by the prediction processing unit of the decoding device. FIG. 44 is a flowchart illustrating another example of the processing performed by the prediction processing unit of the decoding device. FIG. 45 is a flowchart illustrating an example of inter prediction in the normal inter mode in the decoding device. FIG. 46 is a block diagram illustrating an implementation example of a decoding device. FIG. 47 is a flowchart illustrating a first specific example of a decoding process based on the BIO according to the embodiment. FIG. 48 is a conceptual diagram illustrating an example of calculating a horizontal gradient value according to the embodiment. FIG. 49 is a conceptual diagram showing an example of calculating a vertical gradient value in the embodiment. FIG. 50 is a flowchart showing the operation of the encoding device according to the embodiment. FIG. 51 is a flowchart showing an operation of the decoding device according to the embodiment. FIG. 52 is a block diagram illustrating an overall configuration of a content supply system that realizes a content distribution service. FIG. 53 is a conceptual diagram showing an example of an encoding structure at the time of scalable encoding. FIG. 54 is a conceptual diagram illustrating an example of an encoding structure during scalable encoding. FIG. 55 is a conceptual diagram showing a display screen example of a web page. FIG. 56 is a conceptual diagram showing a display screen example of a web page. FIG. 57 is a block diagram illustrating an example of a smartphone. FIG. 58 is a block diagram illustrating a configuration example of a smartphone.

In recent years, it has been studied to encode a moving image using a bi-directional optical flow. The bidirectional optical flow is also called BIO or BDOF. For example, in the bidirectional optical flow, a prediction image in which a prediction value is adjusted in pixel units using a parameter derived based on a pixel value of a reference image in block units and a gradient value of the reference image in block units is obtained. Generated.

The encoding device is likely to be able to generate a highly accurate predicted image using the bidirectional optical flow. Similarly, there is a high possibility that the decoding apparatus can generate a highly accurate predicted image using the bidirectional optical flow.

{For example, the encoding device encodes a difference image between the predicted image and the original image. Then, the decoding device generates the reconstructed image by decoding the difference image and adding the difference image and the prediction image. By using a highly accurate prediction image, it is possible to reduce the code amount of the difference image. That is, there is a high possibility that the code amount of the moving image can be reduced by using the bidirectional optical flow.

On the other hand, the parameters used for the bidirectional optical flow are derived based on the pixel value and the gradient value at each pixel position of the reference image. Therefore, in order to derive the parameters used for the bidirectional optical flow, the amount of calculation may increase due to the calculation performed for each pixel position.

Therefore, for example, an encoding device according to an embodiment of the present disclosure includes a circuit, and a memory connected to the circuit, wherein the circuit operates in a first range including a first reference block of a current block, And a plurality of pixel positions that are commonly and relatively determined for both a second range including a second reference block of the current block, and a plurality of pixel positions in each of the first range and the second range. For each of the plurality of relative pixel positions, the gradient sum is the absolute value of the sum of the gradient value of the relative pixel position in the first range and the gradient value of the relative pixel position in the second range. Deriving an absolute value, deriving, as a first parameter, a sum of a plurality of gradient sum absolute values derived for each of the plurality of relative pixel positions, For each, a pixel difference value that is a difference between the pixel value of the relative pixel position in the first range and the pixel value of the relative pixel position in the second range is derived, For each of the pixel positions, the relative pixel is determined by the sign of the gradient sum, which is the sum of the gradient value of the relative pixel position in the first range and the gradient value of the relative pixel position in the second range. The sign of the pixel difference value derived for the position is inverted or maintained, and the sum of the plurality of pixel difference values derived for each of the plurality of relative pixel positions and the sign of which is inverted or maintained is calculated. Derived as two parameters, a predicted image used for encoding the current block is generated using the first parameter and the second parameter.

Accordingly, the encoding apparatus can reduce substantial multiplication in which the amount of operation is large in the operation performed for each pixel position, and can derive a plurality of parameters for generating a predicted image with a small amount of operation. it can. Therefore, the encoding device can reduce the amount of processing in encoding.

Further, for example, for each of the plurality of relative pixel positions, the gradient value of the relative pixel position in the first range is a horizontal gradient value of the relative pixel position in the first range, and The gradient value of the relative pixel position in the range is a horizontal gradient value of the relative pixel position in the second range.

Thereby, the encoding device can derive a plurality of parameters with a small amount of calculation using the horizontal gradient value.

In addition, for example, the circuit derives the first parameter by an equation (4.1) described later, derives the second parameter by an equation (4.2) described later, Represents a set of pixel positions, [i, j] represents each of the plurality of relative pixel positions, and for each of the plurality of relative pixel positions, I _x ⁰ is the relative position in the first range. Represents the horizontal gradient value of the pixel position, I _x ¹ represents the horizontal gradient value of the relative pixel position in the second range, and I ⁰ represents the pixel value of the relative pixel position in the first range. , ^{I 1,} the represents the pixel value of the relative pixel position in the second _{^{_{range, abs (I x 1 + I}}} x 0) _represents the absolute value of ^{_{^{I x 1 + I x 0,}}} sign (I x 1 + I _x ⁰⁾ is _a positive ^_I ^x 1 ₊ ^I x ⁰ It represents the sign.

Thereby, the encoding device can appropriately derive a plurality of parameters using the pixel value and the horizontal gradient value.

Further, for example, for each of the plurality of relative pixel positions, the gradient value of the relative pixel position in the first range is a vertical gradient value of the relative pixel position in the first range, and The gradient value of the relative pixel position in the range is a vertical gradient value of the relative pixel position in the second range.

Thereby, the encoding device can derive a plurality of parameters with a small amount of calculation using the vertical gradient value.

In addition, for example, the circuit derives the first parameter by an equation (4.5) described later, derives the second parameter by an equation (4.6) described later, Represents a set of pixel positions, [i, j] represents each of the plurality of relative pixel positions, and for each of the plurality of relative pixel positions, I _y ⁰ is the relative position in the first range. Represents the vertical gradient value of the pixel position, I _y ¹ represents the vertical gradient value of the relative pixel position in the second range, and I ⁰ represents the pixel value of the relative pixel position in the first range. , I ¹ represents the pixel value at the relative pixel position in the second range, abs (I _y ¹ + I _y ⁰ ) represents the absolute value of I _y ¹ + I _y ⁰ , and sign (I _y ¹ + I _y ⁰⁾ is _a positive ^_I ^y 1 ₊ ^I y ⁰ It represents the sign.

Thereby, the encoding device can appropriately derive a plurality of parameters using the pixel value and the vertical gradient value.

Further, for example, for each of the plurality of relative pixel positions, the circuit calculates a horizontal gradient value of the relative pixel position in the first range and a horizontal gradient value of the relative pixel position in the second range. Deriving a horizontal gradient sum absolute value that is an absolute value of the sum of the plurality of horizontal gradient sum absolute values respectively derived for the plurality of relative pixel positions, , A vertical gradient sum absolute value that is the absolute value of the sum of the vertical gradient value of the relative pixel position in the first range and the vertical gradient value of the relative pixel position in the second range, Deriving a sum of a plurality of vertical gradient sum absolute values respectively derived for a plurality of relative pixel positions, wherein a sum of the plurality of horizontal gradient sum absolute values is greater than a sum of the plurality of vertical gradient sum absolute values; The horizontal gradient value of the relative pixel position in the first range is applied to the gradient value of the relative pixel position in the first range for each of the plurality of relative pixel positions; A horizontal gradient value of the relative pixel position in the second range is applied to a gradient value of the relative pixel position in the range, and the sum of the plurality of horizontal gradient sum absolute values is the sum of the plurality of vertical gradient sum absolute values. If not greater than the sum, the vertical gradient value of the relative pixel position in the first range is applied to the gradient value of the relative pixel position in the first range for each of the plurality of relative pixel positions. The vertical gradient value of the relative pixel position in the second range is applied to the gradient value of the relative pixel position in the second range.

Thereby, the encoding device can apply the horizontal gradient value or the vertical gradient value to the derivation of the plurality of parameters based on the comparison between the sum related to the horizontal gradient value and the sum related to the vertical gradient value. . That is, the encoding apparatus can apply a gradient value having a large effect to the derivation of a plurality of parameters.

Further, for example, the circuit derives a third parameter by using the first parameter as a denominator and uses the second parameter as a numerator, and generates the prediction image using the third parameter. .

Thereby, the encoding device can appropriately generate a predicted image using the parameters derived based on the relationship between the plurality of parameters.

In addition, for example, for each of the plurality of relative pixel positions, the circuit may use a three-tap filter having a filter coefficient set of [−1, 0, 1] to generate a gradient of the relative pixel position in the first range. A value and a gradient value of the relative pixel position in the second range.

Thereby, the encoding device can simply and appropriately derive the gradient value at each pixel position.

{In addition, for example, the current block is a sub-block defined by a 4 × 4 pixel size in a coding unit, and the circuit derives the first parameter and the second parameter in units of the sub-block.

Thereby, the encoding device can appropriately generate a predicted image using a plurality of parameters determined by the 4 × 4 pixel size.

Further, for example, a decoding device according to an aspect of the present disclosure includes a circuit, and a memory connected to the circuit, wherein the circuit operates in a first range including a first reference block of a current block; A plurality of pixel positions that are commonly and relatively determined with respect to both of a second range including a second reference block of the current block, and a plurality of pixel positions in each of the first range and the second range. For each of a plurality of relative pixel positions, a gradient sum absolute which is an absolute value of a sum of a gradient value of the relative pixel position in the first range and a gradient value of the relative pixel position in the second range. Deriving a sum of a plurality of gradient sum absolute values respectively derived for the plurality of relative pixel positions as a first parameter; For this, a pixel difference value that is a difference between the pixel value of the relative pixel position in the first range and the pixel value of the relative pixel position in the second range is derived, and the plurality of relative pixel positions are derived. For each relative pixel position, the sign of the gradient sum, which is the sum of the gradient value of the relative pixel position in the first range and the gradient value of the relative pixel position in the second range, The sign of the derived pixel difference value is inverted or maintained, and the sum of the plurality of pixel difference values that are respectively derived for the plurality of relative pixel positions and the sign of which is inverted or maintained is defined as a second parameter. And a predicted image used for decoding the current block is generated using the first parameter and the second parameter.

Accordingly, the decoding device can reduce substantial multiplication in which the amount of operation is large in the operation performed for each pixel position, and can derive a plurality of parameters for generating a predicted image with a small amount of operation. . Therefore, the decoding device can reduce the amount of processing in decoding.

Thereby, the decoding device can derive a plurality of parameters with a small amount of calculation using the horizontal gradient value.

Thereby, the decoding device can appropriately derive a plurality of parameters using the pixel value and the horizontal gradient value.

Thereby, the decoding device can derive a plurality of parameters with a small amount of calculation using the vertical gradient value.

Thereby, the decoding device can appropriately derive a plurality of parameters using the pixel value and the vertical gradient value.

Thus, the decoding device can apply the horizontal gradient value or the vertical gradient value to the derivation of the plurality of parameters based on the comparison between the sum related to the horizontal gradient value and the sum related to the vertical gradient value. That is, the decoding device can apply a gradient value having a large effect to the derivation of a plurality of parameters.

により Thereby, the decoding device can appropriately generate a predicted image using the parameters derived based on the relationship between the plurality of parameters.

Thereby, the decoding device can simply and appropriately derive the gradient value at each pixel position.

Thereby, the decoding device can appropriately generate a predicted image using a plurality of parameters determined by the 4 × 4 pixel size.

Also, for example, the encoding method according to an aspect of the present disclosure is common to both the first range including the first reference block of the current block and the second range including the second reference block of the current block. And a plurality of relative pixel positions, which are a plurality of pixel positions relatively determined and are a plurality of pixel positions in each of the first range and the second range. Deriving a gradient sum absolute value, which is an absolute value of the sum of the gradient value of the position and the gradient value of the relative pixel position in the second range, and calculating a plurality of gradients derived for the plurality of relative pixel positions, respectively. Deriving a sum of the sum absolute values as a first parameter, and for each of the plurality of relative pixel positions, a pixel value of the relative pixel position in the first range; Deriving a pixel difference value that is a difference between the pixel value of the relative pixel position in the two ranges and the gradient value of the relative pixel position in the first range for each of the plurality of relative pixel positions; The sign of the pixel difference value derived for the relative pixel position is inverted or maintained by the sign of the gradient sum that is the sum of the gradient value of the relative pixel position in the second range and the plurality of relative values. The second block derives, as a second parameter, a sum of a plurality of pixel difference values, each of which is derived for a target pixel position and whose sign is inverted or maintained, and uses the first parameter and the second parameter to calculate the current block. Generates a predicted image used for encoding of.

Accordingly, it is possible to reduce substantial multiplication in which the calculation amount is large in the calculation performed for each pixel position, and to derive a plurality of parameters for generating a predicted image with a small calculation amount. . Therefore, it is possible to reduce the amount of processing in encoding.

Also, for example, the decoding method according to an aspect of the present disclosure is common to both a first range including a first reference block of a current block and a second range including a second reference block of the current block. For each of a plurality of relative pixel positions that are a plurality of pixel positions that are relatively determined and that are a plurality of pixel positions in each of the first range and the second range, the relative pixel position in the first range And a gradient sum absolute value that is an absolute value of a sum of the gradient value of the relative pixel position in the second range and the gradient value of the relative pixel position in the second range. The sum of absolute values is derived as a first parameter, and for each of the plurality of relative pixel positions, a pixel value of the relative pixel position in the first range and the Deriving a pixel difference value that is a difference between the pixel value of the relative pixel position in the range and the gradient value of the relative pixel position in the first range for each of the plurality of relative pixel positions; Inverting or maintaining the sign of the pixel difference value derived for the relative pixel position by using the sign of the gradient sum that is the sum of the gradient value of the relative pixel position and the gradient value of the relative pixel position in the two ranges, A total sum of a plurality of pixel difference values, each of which is derived for a pixel position and whose sign is inverted or maintained, is derived as a second parameter, and the current block of the current block is derived using the first parameter and the second parameter. Generate a predicted image used for decoding.

Accordingly, it is possible to reduce substantial multiplication in which the calculation amount is large in the calculation performed for each pixel position, and to derive a plurality of parameters for generating a predicted image with a small calculation amount. . Therefore, it is possible to reduce the amount of processing in decoding.

Alternatively, for example, an encoding device according to an aspect of the present disclosure is an encoding device that encodes a moving image using a predicted image, and includes a division unit, an intra prediction unit, an inter prediction unit, and a conversion unit. , A quantization unit, and an entropy encoding unit.

分割 The dividing unit divides a current picture constituting the video into a plurality of blocks. The intra prediction unit performs intra prediction that generates the predicted image of a current block to be coded in the current picture using a reference image in the current picture to be coded. The inter prediction unit performs inter prediction that generates the predicted image of the current block using a reference image in a reference picture different from the current picture.

The conversion unit converts the prediction image generated by the intra prediction unit or the inter prediction unit and a prediction error signal between the image of the encoding target block and transform coefficients of the encoding target block. Generate a signal. The quantization unit quantizes the transform coefficient signal. The entropy encoding unit encodes the quantized transform coefficient signal.

Also, for example, the inter prediction unit is commonly and relatively determined for both a first range including a first reference block of a current block and a second range including a second reference block of the current block. For each of a plurality of relative pixel positions that are a plurality of pixel positions and a plurality of pixel positions in each of the first range and the second range, a gradient value of the relative pixel position in the first range; Deriving a gradient sum absolute value that is an absolute value of a sum of the gradient value of the relative pixel position and the gradient value in the second range, and calculating a sum of a plurality of gradient sum absolute values derived for the plurality of relative pixel positions, respectively. Derived as a first parameter, and for each of the plurality of relative pixel positions, the pixel value of the relative pixel position in the first range and the pixel value of the relative pixel position in the second range. A pixel difference value that is a difference between the pixel value of the relative pixel position and the gradient value of the relative pixel position in the first range for each of the plurality of relative pixel positions. By the sign of the gradient sum that is the sum of the gradient value of the relative pixel position in the range, the sign of the pixel difference value derived for the relative pixel position is inverted or maintained, and the plurality of relative pixels The sum of a plurality of pixel difference values, each of which is derived for a position and the sign of which is inverted or maintained, is derived as a second parameter, and the code of the current block is obtained using the first parameter and the second parameter. Generates a prediction image used for image formation.

Alternatively, for example, a decoding device according to an aspect of the present disclosure is a decoding device that decodes a moving image using a predicted image, and includes an entropy decoding unit, an inverse quantization unit, an inverse transform unit, and an intra prediction unit. , An inter prediction unit, and an addition unit (reconstruction unit).

The entropy decoding unit decodes a quantized transform coefficient signal of a decoding target block in a decoding target picture constituting the video. The inverse quantization unit inversely quantizes the quantized transform coefficient signal. The inverse transform unit inversely transforms the transform coefficient signal to obtain a prediction error signal of the current block.

The intra prediction unit performs intra prediction that generates the predicted image of the current block using a reference image in the current picture. The inter prediction unit performs inter prediction that generates the predicted image of the current block using a reference image in a reference picture different from the current picture. The adding unit reconstructs the image of the decoding target block by adding the prediction image generated by the intra prediction unit or the inter prediction unit and the prediction error signal.

Also, for example, the inter prediction unit is commonly and relatively determined for both a first range including a first reference block of a current block and a second range including a second reference block of the current block. For each of a plurality of relative pixel positions that are a plurality of pixel positions and a plurality of pixel positions in each of the first range and the second range, a gradient value of the relative pixel position in the first range; Deriving a gradient sum absolute value that is an absolute value of a sum of the gradient value of the relative pixel position and the gradient value in the second range, and calculating a sum of a plurality of gradient sum absolute values derived for the plurality of relative pixel positions, respectively. Derived as a first parameter, and for each of the plurality of relative pixel positions, the pixel value of the relative pixel position in the first range and the pixel value of the relative pixel position in the second range. A pixel difference value that is a difference between the pixel value of the relative pixel position and the gradient value of the relative pixel position in the first range for each of the plurality of relative pixel positions. By the sign of the gradient sum that is the sum of the gradient value of the relative pixel position in the range, the sign of the pixel difference value derived for the relative pixel position is inverted or maintained, and the plurality of relative pixels Deriving a sum of a plurality of pixel difference values, each of which is derived for a position and whose sign is inverted or maintained, as a second parameter, and decodes the current block using the first parameter and the second parameter. Generate a predicted image used for.

Furthermore, these generic or specific aspects may be implemented in a non-transitory recording medium such as a system, an apparatus, a method, an integrated circuit, a computer program, or a computer-readable CD-ROM. , An apparatus, a method, an integrated circuit, a computer program, and a recording medium.

Hereinafter, embodiments will be specifically described with reference to the drawings. Each of the embodiments described below shows a comprehensive or specific example. Numerical values, shapes, materials, constituent elements, arrangement positions and connection forms of constituent elements, steps, relations of steps, and orders shown in the following embodiments are merely examples, and do not limit the scope of the claims.

Hereinafter, embodiments of the encoding device and the decoding device will be described. Embodiments are examples of an encoding device and a decoding device to which the processing and / or configuration described in each aspect of the present disclosure can be applied. The processing and / or configuration can be implemented in an encoding device and a decoding device different from those in the embodiment. For example, regarding the processing and / or configuration applied to the embodiment, for example, any of the following may be performed.

(1) Any one of a plurality of components of the encoding device or the decoding device according to an embodiment described in each aspect of the present disclosure is replaced with another component described in any of the aspects of the present disclosure, or They may be combined.

(2) In the encoding device or the decoding device according to the embodiment, addition or replacement of a function or a process is performed with respect to a function or a process performed by some of the components of the encoding device or the decoding device. Any changes, such as deletions, may be made. For example, any function or process may be replaced or combined with another function or process described in any of the aspects of the present disclosure.

(3) In the method performed by the encoding device or the decoding device according to the embodiment, even if an arbitrary change such as addition, replacement, or deletion is made to some of the plurality of processes included in the method. Good. For example, any processing in the method may be replaced or combined with other processing described in any of the aspects of the present disclosure.

(4) Some of the components constituting the encoding device or the decoding device according to the embodiment may be combined with components described in any of the aspects of the present disclosure. May be combined with a component having a part of the function described in any of the aspects of the present disclosure, or a component that performs a part of a process performed by the component described in each of the aspects of the present disclosure May be combined.

(5) A component having a part of the function of the encoding device or the decoding device according to the embodiment, or a component performing a part of the processing of the encoding device or the decoding device according to the embodiment, A component described in any of the aspects, a component having a part of the function described in any of the aspects of the present disclosure, or a part of the processing described in any of the aspects of the present disclosure It may be combined with or replaced by a component to be implemented.

(6) In the method performed by the encoding device or the decoding device according to the embodiment, any one of a plurality of processes included in the method may be a process described in any of the aspects of the present disclosure, or may be a similar process. Any of the processes may be replaced or combined.

(7) Some of the plurality of processes included in the method performed by the encoding device or the decoding device according to the embodiment may be combined with the processes described in any of the aspects of the present disclosure. .

(8) The manner of implementing the processing and / or configuration described in each aspect of the present disclosure is not limited to the encoding device or the decoding device according to the embodiment. For example, the processing and / or the configuration may be performed in an apparatus used for a purpose different from the moving image encoding or the moving image decoding disclosed in the embodiment.

[Encoding device]
First, an encoding device according to an embodiment will be described. FIG. 1 is a block diagram showing a functional configuration of an encoding device 100 according to the embodiment. The encoding device 100 is a moving image encoding device that encodes a moving image in block units.

As shown in FIG. 1, an encoding apparatus 100 is an apparatus that encodes an image in units of blocks, and includes a division unit 102, a subtraction unit 104, a conversion unit 106, a quantization unit 108, and entropy encoding. Unit 110, inverse quantization unit 112, inverse transform unit 114, addition unit 116, block memory 118, loop filter unit 120, frame memory 122, intra prediction unit 124, inter prediction unit 126, And a prediction control unit 128.

The encoding device 100 is realized by, for example, a general-purpose processor and a memory. In this case, when the software program stored in the memory is executed by the processor, the processor includes the dividing unit 102, the subtracting unit 104, the transforming unit 106, the quantizing unit 108, the entropy encoding unit 110, and the inverse quantizing unit 112. , The inverse transform unit 114, the adder unit 116, the loop filter unit 120, the intra prediction unit 124, the inter prediction unit 126, and the prediction control unit 128. The encoding apparatus 100 includes a dividing unit 102, a subtracting unit 104, a transforming unit 106, a quantizing unit 108, an entropy encoding unit 110, an inverse quantizing unit 112, an inverse transforming unit 114, an adding unit 116, and a loop filter unit 120. , The intra prediction unit 124, the inter prediction unit 126, and the prediction control unit 128.

Hereinafter, after describing the overall processing flow of the encoding device 100, each component included in the encoding device 100 will be described.

[Overall flow of encoding process]
FIG. 2 is a flowchart illustrating an example of an overall encoding process performed by the encoding device 100.

First, the dividing unit 102 of the encoding device 100 divides each picture included in an input image that is a moving image into a plurality of fixed-size blocks (for example, 128 × 128 pixels) (Step Sa_1). Then, the division unit 102 selects a division pattern (also referred to as a block shape) for the fixed-size block (Step Sa_2). That is, the dividing unit 102 further divides the fixed-size block into a plurality of blocks constituting the selected division pattern. Then, for each of the plurality of blocks, the encoding device 100 performs the processing of steps Sa_3 to Sa_9 on the block (that is, the encoding target block).

That is, the prediction processing unit including all or a part of the intra prediction unit 124, the inter prediction unit 126, and the prediction control unit 128 generates a prediction signal (also referred to as a prediction block) of the current block (also referred to as a current block). (Step Sa_3).

Next, the subtraction unit 104 generates a difference between the current block and the prediction block as a prediction residual (also referred to as a difference block) (Step Sa_4).

Next, the conversion unit 106 and the quantization unit 108 generate a plurality of quantized coefficients by performing conversion and quantization on the difference block (step Sa_5). Note that a block including a plurality of quantized coefficients is also referred to as a coefficient block.

Next, the entropy coding unit 110 generates a coded signal by performing coding (specifically, entropy coding) on the coefficient block and a prediction parameter related to generation of a prediction signal (step S <b> 1). Sa_6). Note that the encoded signal is also referred to as an encoded bit stream, a compressed bit stream, or a stream.

Next, the inverse quantization unit 112 and the inverse transformation unit 114 restore a plurality of prediction residuals (that is, difference blocks) by performing inverse quantization and inverse transformation on the coefficient block (step Sa_7).

Next, the adding unit 116 reconstructs the current block into a reconstructed image (also referred to as a reconstructed block or a decoded image block) by adding a prediction block to the restored difference block (step Sa_8). As a result, a reconstructed image is generated.

と When the reconstructed image is generated, the loop filter unit 120 performs filtering on the reconstructed image as needed (step Sa_9).

Then, the coding apparatus 100 determines whether or not the coding of the entire picture has been completed (step Sa_10), and when it is determined that the coding has not been completed (No in step Sa_10), the processing from step Sa_2 is repeatedly executed. I do.

In the above-described example, the encoding device 100 selects one division pattern for a fixed-size block and encodes each block according to the division pattern. Each block may be coded. In this case, the encoding device 100 evaluates the cost for each of the plurality of divided patterns, and, for example, uses an encoded signal obtained by encoding according to the divided pattern with the lowest cost as an output encoded signal. You may choose.

よう As shown, the processing of steps Sa_1 to Sa_10 is sequentially performed by the encoding device 100. Alternatively, some of the processes may be performed in parallel, or the order of the processes may be changed.

[Division section]
The division unit 102 divides each picture included in the input moving image into a plurality of blocks, and outputs each block to the subtraction unit 104. For example, the division unit 102 first divides a picture into blocks of a fixed size (for example, 128 × 128). Other fixed block sizes may be employed. This fixed size block may be referred to as a coding tree unit (CTU). The dividing unit 102 divides each of the fixed-size blocks into variable-size (for example, 64 × 64 or less) blocks based on, for example, recursive quadtree and / or binary tree block division. I do. That is, the division unit 102 selects a division pattern. This variable size block may be called a coding unit (CU), a prediction unit (PU), or a transform unit (TU). In various processing examples, CUs, PUs, and TUs do not need to be distinguished, and some or all blocks in a picture may be processing units of the CUs, PUs, and TUs.

FIG. 3 is a conceptual diagram showing an example of block division according to the embodiment. In FIG. 3, a solid line represents a block boundary obtained by dividing a quadtree block, and a broken line represents a block boundary obtained by dividing a binary tree block.

Here, the block 10 is a square block of 128 × 128 pixels (128 × 128 block). The 128 × 128 block 10 is first divided into four square 64 × 64 blocks (quad tree block division).

６４ The upper left 64 × 64 block is further vertically divided into two rectangular 32 × 64 blocks, and the left 32 × 64 block is further vertically divided into two rectangular 16 × 64 blocks (binary tree block division). As a result, the upper left 64 × 64 block is divided into two 16 × 64

blocks

11 and 12 and a 32 × 64 block 13.

６４ The upper right 64 × 64 block is horizontally divided into two rectangular 64 × 32 blocks 14 and 15 (binary tree block division).

６４ The lower left 64 × 64 block is divided into four square 32 × 32 blocks (quad tree block division). The upper left block and the lower right block of the four 32 × 32 blocks are further divided. The upper left 32 × 32 block is vertically divided into two rectangular 16 × 32 blocks, and the right 16 × 32 block is further horizontally divided into two 16 × 16 blocks (binary tree block division). The lower right 32 × 32 block is horizontally divided into two 32 × 16 blocks (binary tree block division). As a result, the lower left 64x64 block is divided into a 16x32 block 16, two 16x16 blocks 17,18, two 32x32 blocks 19,20, and two 32x16 blocks 21,22.

６４ The lower right 64 × 64 block 23 is not divided.

As described above, in FIG. 3, the block 10 is divided into thirteen variable-size blocks 11 to 23 based on recursive quadtree and binary tree block division. Such division may be referred to as QTBT (quad-tree \ plus \ binary \ tree) division.

In FIG. 3, one block is divided into four or two blocks (quadtree or binary tree block division), but the division is not limited to these. For example, one block may be divided into three blocks (triple tree block division). A division including such a ternary tree block division may be referred to as MBT (multiｍtype tree) division.

[Picture composition slice / tile]
In order to decode pictures in parallel, the pictures may be configured in slice units or tile units. A picture composed of slice units or tile units may be configured by the division unit 102.

Slice is a basic unit of coding that constitutes a picture. A picture is composed of, for example, one or more slices. A slice is composed of one or more continuous CTUs (Coding Tree Units).

FIG. 4A is a conceptual diagram showing an example of the configuration of a slice. For example, a picture includes 11 × 8 CTUs and is divided into four slices (slices 1-4). Slice 1 is composed of 16 CTUs, slice 2 is composed of 21 CTUs, slice 3 is composed of 29 CTUs, and slice 4 is composed of 22 CTUs. Here, each CTU in the picture belongs to one of the slices. The shape of the slice is a shape obtained by dividing the picture in the horizontal direction. The boundary of the slice does not need to be the edge of the screen, and may be any of the boundaries of the CTU in the screen. The processing order (encoding order or decoding order) of the CTU in the slice is, for example, a raster scan order. Each slice includes header information and encoded data. The header information may describe characteristics of the slice, such as the CTU address at the head of the slice and the slice type.

A tile is a unit of a rectangular area constituting a picture. A number called TileId may be assigned to each tile in raster scan order.

FIG. 4B is a conceptual diagram showing an example of the configuration of a tile. For example, a picture includes 11 × 8 CTUs and is divided into four rectangular area tiles (tiles 1-4). When the tile is used, the processing order of the CTU is changed as compared with the case where the tile is not used. If no tiles are used, the CTUs in the picture are processed in raster scan order. If tiles are used, at least one CTU in each of the plurality of tiles is processed in raster scan order. For example, as shown in FIG. 4B, the processing order of a plurality of CTUs included in tile 1 is from the left end of the first row of tile 1 to the right end of the first row of tile 1, and then the left end of the second row of tile 1 To the right end of the second row of the tile 1.

Note that one tile may include one or more slices, and one slice may include one or more tiles.

[Subtraction unit]
The subtraction unit 104 subtracts a prediction signal (a prediction sample input from the prediction control unit 128 shown below) from an original signal (original sample) in block units input from the division unit 102 and divided by the division unit 102. . That is, the subtraction unit 104 calculates a prediction error (also referred to as a residual) of the current block (hereinafter, referred to as a current block). Then, the subtraction unit 104 outputs the calculated prediction error (residual error) to the conversion unit 106.

The original signal is an input signal of the encoding apparatus 100, and is a signal (for example, a luminance (luma) signal and two color difference (chroma) signals) representing an image of each picture constituting a moving image. Hereinafter, a signal representing an image may be referred to as a sample.

[Conversion unit]
Transform section 106 transforms the prediction error in the spatial domain into transform coefficients in the frequency domain, and outputs the transform coefficients to quantization section 108. Specifically, the transform unit 106 performs, for example, a predetermined discrete cosine transform (DCT) or a discrete sine transform (DST) on the prediction error in the spatial domain. The predetermined DCT or DST may be predetermined.

Note that the conversion unit 106 adaptively selects a conversion type from a plurality of conversion types, and converts the prediction error into a conversion coefficient using a conversion basis function (transform basis function) corresponding to the selected conversion type. May be. Such a conversion is sometimes called EMT (explicit multiple core transform) or AMT (adaptive multiple multiple transform).

変換 The plurality of conversion types include, for example, DCT-II, DCT-V, DCT-VIII, DST-I and DST-VII. FIG. 5A is a table showing conversion basis functions corresponding to conversion type examples. In FIG. 5A, N indicates the number of input pixels. Selection of a conversion type from among the plurality of conversion types may depend on, for example, the type of prediction (intra prediction and inter prediction) or may depend on the intra prediction mode.

The information indicating whether to apply such EMT or AMT (for example, referred to as an EMT flag or an AMT flag) and the information indicating the selected conversion type are usually signaled at the CU level. Note that the signalization of these pieces of information need not be limited to the CU level, but may be another level (for example, a bit sequence level, a picture level, a slice level, a tile level, or a CTU level).

(4) The conversion unit 106 may re-convert the conversion coefficient (conversion result). Such re-transformation may be referred to as AST (adaptive @ secondary @ transform) or NSST (non-separable @ secondary @ transform). For example, the transform unit 106 performs re-conversion for each sub-block (for example, a 4 × 4 sub-block) included in a block of a transform coefficient corresponding to an intra prediction error. The information indicating whether to apply the NSST and the information on the transformation matrix used for the NSST are usually signaled at the CU level. The signalization of these pieces of information need not be limited to the CU level, but may be another level (for example, a sequence level, a picture level, a slice level, a tile level, or a CTU level).

Separable conversion and Non-Separable conversion may be applied to the conversion unit 106. Separable conversion is a method of performing conversion a plurality of times by separating each direction by the number of input dimensions. Non-separable conversion is a method of converting two or more dimensions when the input is multidimensional. This is a method in which conversion is performed collectively assuming that the data is one-dimensional.

For example, as an example of a non-separable conversion, if an input is a 4 × 4 block, it is regarded as one array having 16 elements, and a 16 × 16 conversion matrix is applied to the array. , Which performs the conversion process.

In a further example of the non-separable conversion, a conversion (Hypercube) in which a 4 × 4 input block is regarded as one array having 16 elements, and a Givens rotation is performed on the array a plurality of times (Hypercube). Gives @ Transform) may be performed.

変換 In the conversion by the conversion unit 106, the type of base to be converted to the frequency domain can be switched according to the area in the CU. As an example, there is SVT (Spatially Varying Transform). In the SVT, as shown in FIG. 5B, the CU is divided into two equal parts in the horizontal or vertical direction, and only one of the areas is converted into the frequency area. The type of the transformation base can be set for each area, and for example, DST7 and DCT8 are used. In this example, only one of the two areas in the CU is converted and the other is not converted, but both areas may be converted. Further, the dividing method can be made more flexible, such as not only dividing into two, but also dividing into four, or information indicating the division is separately encoded and signaled similarly to the CU division. Note that the SVT may be referred to as SBT (Sub-block @ Transform).

[Quantizer]
The quantization unit 108 quantizes the transform coefficient output from the transform unit 106. Specifically, the quantization unit 108 scans the transform coefficients of the current block in a predetermined scanning order, and quantizes the transform coefficients based on the quantization parameter (QP) corresponding to the scanned transform coefficients. Then, the quantization unit 108 outputs the quantized transform coefficients of the current block (hereinafter, referred to as quantization coefficients) to the entropy encoding unit 110 and the inverse quantization unit 112. The predetermined scanning order may be determined in advance.

The predetermined scanning order is an order for quantization / inverse quantization of transform coefficients. For example, the predetermined scanning order may be defined in ascending frequency order (low-frequency to high-frequency) or descending order (high-frequency to low-frequency).

The quantization parameter (QP) is a parameter that defines a quantization step (quantization width). For example, as the value of the quantization parameter increases, the quantization step also increases. That is, as the value of the quantization parameter increases, the quantization error increases.

In some cases, a quantization matrix is used for quantization. For example, several types of quantization matrices may be used in correspondence with frequency transform sizes such as 4x4 and 8x8, prediction modes such as intra prediction and inter prediction, and pixel components such as luminance and color difference. Note that quantization refers to digitizing a value sampled at a predetermined interval in association with a predetermined level, and in this technical field, is referred to using other expressions such as rounding, rounding, and scaling. Alternatively, rounding, rounding, and scaling may be employed. The predetermined interval and level may be determined in advance.

As a method of using a quantization matrix, there are a method of using a quantization matrix directly set on the encoding device side and a method of using a default quantization matrix (default matrix). On the encoding device side, by directly setting the quantization matrix, it is possible to set the quantization matrix according to the characteristics of the image. However, in this case, there is a disadvantage that the coding amount is increased by coding the quantization matrix.

On the other hand, there is a method in which the coefficient of the high-frequency component and the coefficient of the low-frequency component are similarly quantized without using the quantization matrix. This method is equivalent to a method using a quantization matrix (flat matrix) in which all coefficients have the same value.

The マトリックス quantization matrix may be specified by, for example, SPS (Sequence Parameter Set: Sequence Parameter Set) or PPS (Picture Parameter Set: Picture Parameter Set). The SPS includes parameters used for sequences, and the PPS includes parameters used for pictures. SPS and PPS may be simply referred to as a parameter set.

[Entropy encoder]
The entropy coding unit 110 generates a coded signal (coded bit stream) based on the quantized coefficients input from the quantization unit 108. Specifically, for example, the entropy encoding unit 110 binarizes the quantized coefficients, arithmetically encodes the binary signal, and outputs a compressed bit stream or sequence.

[Inverse quantization unit]
The inverse quantization unit 112 inversely quantizes the quantization coefficient input from the quantization unit 108. Specifically, the inverse quantization unit 112 inversely quantizes the quantization coefficient of the current block in a predetermined scanning order. Then, the inverse quantization unit 112 outputs the inversely quantized transform coefficient of the current block to the inverse transformation unit 114. The predetermined scanning order may be determined in advance.

[Inverse conversion unit]
The inverse transform unit 114 restores a prediction error (residual error) by inversely transforming the transform coefficient input from the inverse quantization unit 112. Specifically, the inverse transform unit 114 restores the prediction error of the current block by performing an inverse transform corresponding to the transform by the transform unit 106 on the transform coefficient. Then, the inverse transform unit 114 outputs the restored prediction error to the adding unit 116.

Note that the restored prediction error usually does not match the prediction error calculated by the subtraction unit 104 because information is lost due to quantization. That is, the restored prediction error usually includes a quantization error.

[Adder]
The addition unit 116 reconstructs the current block by adding the prediction error input from the inverse transform unit 114 and the prediction sample input from the prediction control unit 128. Then, the adding unit 116 outputs the reconstructed block to the block memory 118 and the loop filter unit 120. The reconstructed block is sometimes called a local decoding block.

[Block memory]
The block memory 118 is, for example, a storage unit for storing a block that is referred to in intra prediction and is in a current picture to be coded (called a current picture). Specifically, the block memory 118 stores the reconstructed block output from the adding unit 116.

[Frame memory]
The frame memory 122 is, for example, a storage unit for storing reference pictures used for inter prediction, and may be called a frame buffer. Specifically, the frame memory 122 stores the reconstructed blocks filtered by the loop filter unit 120.

[Loop filter section]
The loop filter unit 120 applies a loop filter to the block reconstructed by the adding unit 116, and outputs the reconstructed block that has been filtered to the frame memory 122. The loop filter is a filter (in-loop filter) used in the encoding loop, and includes, for example, a deblocking filter (DF or DBF), a sample adaptive offset (SAO), an adaptive loop filter (ALF), and the like.

In ALF, a least squares error filter for removing coding distortion is applied. For example, for every 2 × 2 sub-block in the current block, a plurality of sub-blocks are determined based on the direction and activity of a local gradient. One filter selected from the filters is applied.

{Specifically, first, sub-blocks (for example, 2 × 2 sub-blocks) are classified into a plurality of classes (for example, 15 or 25 classes). The classification of the sub-blocks is performed based on the direction and the activity of the gradient. For example, the classification value C (for example, C = 5D + A) is calculated using the gradient direction value D (for example, 0 to 2 or 0 to 4) and the gradient activity value A (for example, 0 to 4). Then, the sub-blocks are classified into a plurality of classes based on the classification value C.

The gradient direction value D is derived, for example, by comparing gradients in a plurality of directions (for example, horizontal, vertical and two diagonal directions). The gradient activation value A is derived, for example, by adding gradients in a plurality of directions and quantizing the addition result.

フィルタ A filter for a sub-block is determined from a plurality of filters based on the result of such classification.

As the shape of the filter used in the ALF, for example, a circularly symmetric shape is used. 6A to 6C are views showing a plurality of examples of the shape of the filter used in the ALF. 6A shows a 5 × 5 diamond-shaped filter, FIG. 6B shows a 7 × 7 diamond-shaped filter, and FIG. 6C shows a 9 × 9 diamond-shaped filter. The information indicating the shape of the filter is usually signaled at the picture level. The signalization of the information indicating the shape of the filter need not be limited to the picture level, but may be another level (for example, a sequence level, a slice level, a tile level, a CTU level, or a CU level).

$ ON / OFF of ALF may be determined, for example, at a picture level or a CU level. For example, whether to apply ALF at the CU level may be determined for luminance, and whether to apply ALF at the picture level may be determined for color difference. The information indicating ALF on / off is usually signaled at a picture level or a CU level. The signalization of the information indicating ON / OFF of the ALF does not need to be limited to the picture level or the CU level, and may be at another level (for example, a sequence level, a slice level, a tile level, or a CTU level). Good.

The set of coefficients for a plurality of selectable filters (eg up to 15 or 25 filters) is usually signaled at the picture level. The signalization of the coefficient set need not be limited to the picture level, but may be another level (for example, a sequence level, a slice level, a tile level, a CTU level, a CU level, or a sub-block level).

[Loop filter> Deblocking filter]
In the deblocking filter, the loop filter unit 120 performs a filtering process on a block boundary of a reconstructed image to reduce distortion generated at the block boundary.

FIG. 7 is a block diagram showing an example of a detailed configuration of the loop filter unit 120 functioning as a deblocking filter.

The loop filter unit 120 includes a boundary determination unit 1201, a filter determination unit 1203, a filter processing unit 1205, a processing determination unit 1208, a filter characteristic determination unit 1207, and switches 1202, 1204, and 1206.

The boundary determination unit 1201 determines whether or not a pixel to be subjected to deblocking filtering (that is, a target pixel) exists near a block boundary. Then, boundary determination section 1201 outputs the determination result to switch 1202 and processing determination section 1208.

When the boundary determination unit 1201 determines that the target pixel exists near the block boundary, the switch 1202 outputs the image before the filter processing to the switch 1204. Conversely, when the boundary determination unit 1201 determines that the target pixel does not exist near the block boundary, the switch 1202 outputs the image before the filter processing to the switch 1206.

The filter determination unit 1203 determines whether to perform the deblocking filter processing on the target pixel based on the pixel values of at least one peripheral pixel around the target pixel. Then, filter determination section 1203 outputs the determination result to switch 1204 and processing determination section 1208.

If the filter determination unit 1203 determines that the deblocking filter processing is to be performed on the target pixel, the switch 1204 outputs the image before the filter processing obtained via the switch 1202 to the filter processing unit 1205. Conversely, when the filter determination unit 1203 determines that the deblocking filter processing is not performed on the target pixel, the switch 1204 outputs the image before the filter processing acquired via the switch 1202 to the switch 1206.

When acquiring the image before the filter processing via the

switches

1202 and 1204, the filter processing unit 1205 performs the deblocking filter processing having the filter characteristics determined by the filter characteristic determination unit 1207 on the target pixel. Execute. Then, the filter processing unit 1205 outputs the pixel after the filter processing to the switch 1206.

The switch 1206 selectively outputs a pixel that has not been deblocking-filtered and a pixel that has been deblocking-filtered by the filter processing unit 1205 under the control of the processing determination unit 1208.

The processing determination unit 1208 controls the switch 1206 based on the determination results of the boundary determination unit 1201 and the filter determination unit 1203. That is, when the processing determination unit 1208 determines that the target pixel exists near the block boundary by the boundary determination unit 1201 and determines that the filter determination unit 1203 performs the deblocking filter processing on the target pixel. , The pixel subjected to the deblocking filter processing is output from the switch 1206. In cases other than those described above, the processing determining unit 1208 causes the switch 1206 to output a pixel that has not been subjected to the deblocking filter processing. By repeatedly outputting such pixels, the image after the filter processing is output from the switch 1206.

FIG. 8 is a conceptual diagram showing an example of a deblocking filter having filter characteristics symmetric with respect to a block boundary.

In the deblocking filter processing, for example, one of two deblocking filters having different characteristics, that is, a strong filter and a weak filter is selected using a pixel value and a quantization parameter. In the strong filter, as shown in FIG. 8, when pixels p0 to p2 and pixels q0 to q2 are present across a block boundary, the pixel values of the pixels q0 to q2 are calculated, for example, by the following equation. Are performed, the pixel values are changed to pixel values q′0 to q′2.

q′0 = (p1 + 2 × p0 + 2 × q0 + 2 × q1 + q2 + 4) / 8
q′1 = (p0 + q0 + q1 + q2 + 2) / 4
q′2 = (p0 + q0 + q1 + 3 × q2 + 2 × q3 + 4) / 8

In the above equation, p0 to p2 and q0 to q2 are the pixel values of pixels p0 to p2 and pixels q0 to q2, respectively. Further, q3 is a pixel value of the pixel q3 adjacent to the pixel q2 on the opposite side to the block boundary. Further, on the right side of each of the above equations, a coefficient by which the pixel value of each pixel used in the deblocking filter processing is multiplied is a filter coefficient.

Furthermore, in the deblocking filter processing, clip processing may be performed so that the pixel value after calculation is not set to exceed the threshold value. In this clipping process, the pixel value after the calculation according to the above equation is clipped to “calculation target pixel value ± 2 × threshold” using the threshold value determined from the quantization parameter. Thereby, excessive smoothing can be prevented.

FIG. 9 is a conceptual diagram for explaining a block boundary where deblocking filter processing is performed. FIG. 10 is a conceptual diagram illustrating an example of the Bs value.

The block boundary where the deblocking filter processing is performed is, for example, a boundary of a PU (Prediction @ Unit) or a TU (Transform @ Unit) of an 8 × 8 pixel block as shown in FIG. The deblocking filtering may be performed in units of four rows or four columns. First, a Bs (Boundary Strength) value is determined for the blocks P and Q shown in FIG. 9 as shown in FIG.

According to the Bs value in FIG. 10, whether or not to perform deblocking filter processing of different strength is determined even for block boundaries belonging to the same image. The deblocking filter processing on the color difference signal is performed when the Bs value is 2. The deblocking filter processing on the luminance signal is performed when the Bs value is 1 or more and a predetermined condition is satisfied. The predetermined condition may be determined in advance. Note that the determination condition of the Bs value is not limited to the one shown in FIG. 10 and may be determined based on another parameter.

[Prediction processing unit (intra prediction unit / inter prediction unit / prediction control unit)]
FIG. 11 is a flowchart illustrating an example of processing performed by the prediction processing unit of the encoding device 100. The prediction processing unit includes all or some components of the intra prediction unit 124, the inter prediction unit 126, and the prediction control unit 128.

The prediction processing unit generates a predicted image of the current block (Step Sb_1). This prediction image is also called a prediction signal or a prediction block. The prediction signal includes, for example, an intra prediction signal or an inter prediction signal. Specifically, the prediction processing unit generates a reconstructed image that has already been obtained by performing generation of a prediction block, generation of a difference block, generation of a coefficient block, restoration of a difference block, and generation of a decoded image block. To generate a predicted image of the current block.

The reconstructed image may be, for example, an image of a reference picture or an image of a coded block in the current picture which is a picture including the current block. The coded block in the current picture is, for example, a block adjacent to the current block.

FIG. 12 is a flowchart illustrating another example of the processing performed by the prediction processing unit of the encoding device 100.

The prediction processing unit generates a predicted image using the first method (Step Sc_1a), generates a predicted image using the second method (Step Sc_1b), and generates a predicted image using the third method (Step Sc_1c). The first scheme, the second scheme, and the third scheme are different schemes for generating a predicted image, and are, for example, inter prediction schemes, intra prediction schemes, and other prediction schemes, respectively. There may be. In these prediction methods, the above-described reconstructed image may be used.

Next, the prediction processing unit selects one of the plurality of predicted images generated in steps Sc_1a, Sc_1b, and Sc_1c (step Sc_2). The selection of the predicted image, that is, the selection of a method or a mode for obtaining a final predicted image may be performed based on the calculated cost for each generated predicted image. Alternatively, the selection of the predicted image may be performed based on parameters used for the encoding process. The encoding device 100 may signal information for specifying the selected predicted image, scheme, or mode into an encoded signal (also referred to as an encoded bit stream). The information may be, for example, a flag. Thereby, the decoding device can generate a predicted image according to the method or mode selected in encoding device 100 based on the information. In the example illustrated in FIG. 12, the prediction processing unit selects one of the predicted images after generating the predicted image in each method. However, the prediction processing unit, before generating those predicted images, based on the parameters used in the above-described encoding processing, select a method or mode, and generate a predicted image according to the method or mode Is also good.

For example, the first method and the second method are intra prediction and inter prediction, respectively, and the prediction processing unit generates a final prediction image for the current block from prediction images generated according to these prediction methods. You may choose.

FIG. 13 is a flowchart illustrating another example of the processing performed by the prediction processing unit of the encoding device 100.

First, the prediction processing unit generates a predicted image by intra prediction (step Sd_1a), and generates a predicted image by inter prediction (step Sd_1b). Note that a predicted image generated by intra prediction is also called an intra predicted image, and a predicted image generated by inter prediction is also called an inter predicted image.

Next, the prediction processing unit evaluates each of the intra prediction image and the inter prediction image (Step Sd_2). Cost may be used for this evaluation. That is, the prediction processing unit calculates the respective costs C of the intra prediction image and the inter prediction image. This cost C can be calculated by the formula of the RD optimization model, for example, C = D + λ × R. In this equation, D is the encoding distortion of the predicted image, and is represented by, for example, the sum of absolute differences between the pixel value of the current block and the pixel value of the predicted image. R is the amount of generated code of the predicted image, and specifically, is the amount of code required for encoding motion information and the like for generating the predicted image. Λ is, for example, an undetermined Lagrange multiplier.

{Then, the prediction processing unit selects, from the intra-predicted image and the inter-predicted image, the predicted image with the smallest cost C calculated as the final predicted image of the current block (Step Sd_3). That is, a prediction method or mode for generating a prediction image of the current block is selected.

[Intra prediction unit]
The intra prediction unit 124 generates a prediction signal (intra prediction signal) by performing intra prediction (also referred to as intra prediction) of the current block with reference to a block in the current picture stored in the block memory 118. Specifically, the intra prediction unit 124 generates an intra prediction signal by performing intra prediction with reference to a sample (for example, a luminance value and a color difference value) of a block adjacent to the current block, and performs prediction control on the intra prediction signal. Output to the unit 128.

{For example, the intra prediction unit 124 performs intra prediction using one of a plurality of prescribed intra prediction modes. The plurality of intra prediction modes usually includes one or more non-directional prediction modes and a plurality of directional prediction modes. The plurality of prescribed modes may be prescribed in advance.

The one or more non-directional prediction modes are, for example, H.264. It includes a Planar prediction mode and a DC prediction mode defined by the H.265 / HEVC standard.

The plurality of direction prediction modes are, for example, H.264. Includes a prediction mode in 33 directions defined by the H.265 / HEVC standard. Note that the plurality of directional prediction modes may further include 32 directional prediction modes (total of 65 directional prediction modes) in addition to the 33 directions. FIG. 14 is a conceptual diagram showing a total of 67 intra prediction modes (two non-directional prediction modes and 65 directional prediction modes) that can be used in intra prediction. Solid arrows indicate H.E. The H.265 / HEVC standard represents 33 directions, and the dashed arrows represent the added 32 directions (two non-directional prediction modes are not shown in FIG. 14).

In various processing examples, a luminance block may be referred to in intra prediction of a chrominance block. That is, the color difference component of the current block may be predicted based on the luminance component of the current block. Such intra prediction is sometimes called CCLM (cross-component @ linear @ model) prediction. Such an intra prediction mode of a chrominance block that refers to a luminance block (for example, referred to as a CCLM mode) may be added as one of the intra prediction modes of a chrominance block.

The intra prediction unit 124 may correct the pixel value after intra prediction based on the gradient of the reference pixel in the horizontal / vertical direction. Intra prediction with such a correction is sometimes called PDPC (position \ dependent \ intra \ prediction \ combination). Information indicating whether or not PDPC is applied (for example, called a PDPC flag) is usually signaled at the CU level. The signalization of this information need not be limited to the CU level, but may be another level (for example, a sequence level, a picture level, a slice level, a tile level, or a CTU level).

[Inter prediction unit]
The inter prediction unit 126 performs inter prediction (also referred to as inter-screen prediction) of the current block with reference to a reference picture stored in the frame memory 122 and being different from the current picture, thereby obtaining a prediction signal (inter prediction). A prediction signal). The inter prediction is performed in units of a current block or a current sub-block (for example, 4 × 4 block) in the current block. For example, the inter prediction unit 126 performs motion estimation on the current block or the current sub-block in the reference picture, and finds a reference block or a sub-block that best matches the current block or the current sub-block. Then, the inter prediction unit 126 acquires motion information (for example, a motion vector) that compensates for a motion or change from the reference block or the sub-block to the current block or the sub-block. The inter prediction unit 126 performs motion compensation (or motion prediction) based on the motion information, and generates an inter prediction signal of a current block or a sub block. The inter prediction unit 126 outputs the generated inter prediction signal to the prediction control unit 128.

The motion information used for motion compensation may be signaled as an inter prediction signal in various forms. For example, a motion vector may be signalized. As another example, a difference between a motion vector and a predicted motion vector (motion \ vector \ predictor) may be signalized.

[Basic flow of inter prediction]
FIG. 15 is a flowchart illustrating an example of a basic flow of inter prediction.

The inter prediction unit 126 first generates a predicted image (Steps Se_1 to Se_3). Next, the subtraction unit 104 generates a difference between the current block and the predicted image as a prediction residual (Step Se_4).

Here, in generating the predicted image, the inter prediction unit 126 determines the motion vector (MV) of the current block (Steps Se_1 and Se_2) and performs motion compensation (Step Se_3) to generate the predicted image. I do. In determining the MV, the inter prediction unit 126 determines the MV by selecting a candidate motion vector (candidate MV) (Step Se_1) and deriving the MV (Step Se_2). The selection of the candidate MV is performed, for example, by selecting at least one candidate MV from the candidate MV list. In deriving the MV, the inter prediction unit 126 selects at least one candidate MV from the at least one candidate MV, and determines the selected at least one candidate MV as the MV of the current block. You may. Alternatively, for each of the at least one selected candidate MV, the inter prediction unit 126 may determine the MV of the current block by searching for a region of a reference picture indicated by the candidate MV. Note that searching for the area of the reference picture may be referred to as motion search (motion @ estimation).

In addition, in the above example, steps Se_1 to Se_3 are performed by the inter prediction unit 126. However, for example, processing such as step Se_1 or step Se_2 may be performed by other components included in the encoding device 100. .

[Motion vector derivation flow]
FIG. 16 is a flowchart illustrating an example of motion vector derivation.

The inter prediction unit 126 derives the MV of the current block in a mode for encoding motion information (for example, MV). In this case, for example, the motion information is encoded as a prediction parameter and signalized. That is, encoded motion information is included in an encoded signal (also referred to as an encoded bit stream).

Alternatively, the inter prediction unit 126 derives the MV in a mode in which motion information is not encoded. In this case, the motion information is not included in the encoded signal.

Here, the MV derivation mode may include a normal inter mode, a merge mode, a FRUC mode, an affine mode, and the like, which will be described later. Among these modes, modes for encoding motion information include a normal inter mode, a merge mode, and an affine mode (specifically, an affine inter mode and an affine merge mode). The motion information may include not only MV but also predicted motion vector selection information described later. The mode in which motion information is not encoded includes a FRUC mode and the like. The inter prediction unit 126 selects a mode for deriving the MV of the current block from the plurality of modes, and derives the MV of the current block using the selected mode.

FIG. 17 is a flowchart showing another example of deriving a motion vector.

The inter prediction unit 126 derives the MV of the current block in a mode for encoding the difference MV. In this case, for example, the difference MV is encoded as a prediction parameter and signalized. That is, the encoded difference MV is included in the encoded signal. The difference MV is a difference between the MV of the current block and the predicted MV.

{Alternatively, the inter prediction unit 126 derives the MV in a mode in which the difference MV is not encoded. In this case, the encoded difference MV is not included in the encoded signal.

Here, as described above, the modes for deriving the MV include a normal inter, a merge mode, a FRUC mode, and an affine mode described later. Among these modes, the modes for encoding the differential MV include a normal inter mode and an affine mode (specifically, an affine inter mode). Modes in which the difference MV is not encoded include a FRUC mode, a merge mode, and an affine mode (specifically, an affine merge mode). The inter prediction unit 126 selects a mode for deriving the MV of the current block from the plurality of modes, and derives the MV of the current block using the selected mode.

[Motion vector derivation flow]
FIG. 18 is a flowchart illustrating another example of deriving a motion vector. There are a plurality of modes in the MV derivation mode, that is, the inter prediction mode. The modes are roughly classified into a mode in which the differential MV is encoded and a mode in which the differential motion vector is not encoded. The modes in which the difference MV is not encoded include a merge mode, a FRUC mode, and an affine mode (specifically, an affine merge mode). The details of these modes will be described later, but for simplicity, the merge mode is a mode in which the MV of the current block is derived by selecting a motion vector from surrounding encoded blocks, and the FRUC mode is In this mode, the MV of the current block is derived by performing a search between encoded regions. The affine mode is a mode in which a motion vector of each of a plurality of sub-blocks constituting a current block is derived as an MV of the current block, assuming an affine transformation.

Specifically, as illustrated, when the inter prediction mode information indicates 0 (0 in Sf_1), the inter prediction unit 126 derives a motion vector in the merge mode (Sf_2). Further, when the inter prediction mode information indicates 1 (1 in Sf_1), the inter prediction unit 126 derives a motion vector in the FRUC mode (Sf_3). When the inter prediction mode information indicates 2 (2 in Sf_1), the inter prediction unit 126 derives a motion vector in an affine mode (specifically, an affine merge mode) (Sf_4). In addition, when the inter prediction mode information indicates 3 (3 in Sf_1), the inter prediction unit 126 derives a motion vector in a mode for encoding the difference MV (for example, a normal inter mode) (Sf_5).

[MV derivation> Normal inter mode]
The normal inter mode is an inter prediction mode that derives the MV of the current block from the region of the reference picture indicated by the candidate MV based on a block similar to the image of the current block. In the normal inter mode, the difference MV is encoded.

FIG. 19 is a flowchart showing an example of inter prediction in the normal inter mode.

First, the inter prediction unit 126 acquires a plurality of candidate MVs for the current block based on information such as the MVs of a plurality of encoded blocks around the current block in time or space (step). Sg_1). That is, the inter prediction unit 126 creates a candidate MV list.

Next, the inter prediction unit 126 assigns each of N (N is an integer of 2 or more) candidate MVs out of the plurality of candidate MVs obtained in step Sg_1 to a predicted motion vector candidate (also referred to as a predicted MV candidate). Are extracted according to a predetermined priority (step Sg_2). Note that the priority order may be predetermined for each of the N candidate MVs.

Next, the inter prediction unit 126 selects one predicted motion vector candidate from the N predicted motion vector candidates as a predicted motion vector (also referred to as predicted MV) of the current block (step Sg_3). At this time, the inter prediction unit 126 encodes prediction motion vector selection information for identifying the selected prediction motion vector into a stream. Note that the stream is the above-described coded signal or coded bit stream.

Next, the inter prediction unit 126 derives the MV of the current block with reference to the encoded reference picture (Step Sg_4). At this time, the inter prediction unit 126 further encodes a difference value between the derived MV and the predicted motion vector into a stream as a difference MV. Note that an encoded reference picture is a picture composed of a plurality of blocks reconstructed after encoding.

Finally, the inter prediction unit 126 generates a predicted image of the current block by performing motion compensation on the current block using the derived MV and the encoded reference picture (step Sg_5). Note that the prediction image is the above-described inter prediction signal.

情報 In addition, information indicating the inter prediction mode (normal inter mode in the above example) used for generating the predicted image, which is included in the coded signal, is coded as a prediction parameter, for example.

Note that the candidate MV list may be used in common with lists used in other modes. Further, the process regarding the candidate MV list may be applied to the process regarding a list used in another mode. The process regarding the candidate MV list includes, for example, extraction or selection of the candidate MV from the candidate MV list, rearrangement of the candidate MV, or deletion of the candidate MV.

[MV derivation> merge mode]
The merge mode is an inter prediction mode in which a candidate MV is selected from the candidate MV list as the MV of the current block to derive the MV.

FIG. 20 is a flowchart showing an example of inter prediction in the merge mode.

First, the inter prediction unit 126 acquires a plurality of candidate MVs for the current block based on information such as the MVs of a plurality of encoded blocks around the current block in time or space (step). Sh_1). That is, the inter prediction unit 126 creates a candidate MV list.

Next, the inter prediction unit 126 derives the MV of the current block by selecting one candidate MV from the plurality of candidate MVs acquired in Step Sh_1 (Step Sh_2). At this time, the inter prediction unit 126 encodes MV selection information for identifying the selected candidate MV into a stream.

Finally, the inter prediction unit 126 generates a predicted image of the current block by performing motion compensation on the current block using the derived MV and the encoded reference picture (step Sh_3).

情報 In addition, information indicating the inter prediction mode (merged mode in the above example) used for generating the predicted image, which is included in the coded signal, is coded, for example, as a prediction parameter.

FIG. 21 is a conceptual diagram illustrating an example of a motion vector derivation process of the current picture in the merge mode.

First, a predicted MV list in which predicted MV candidates are registered is generated. As candidates for the prediction MV, spatial adjacent prediction MV, which is the MV of a plurality of encoded blocks spatially located around the target block, and a nearby block that projects the position of the target block in the encoded reference picture There are a temporally adjacent prediction MV, which is an MV possessed, a combined prediction MV, which is an MV generated by combining the MV values of the spatially adjacent prediction MV and the temporally adjacent prediction MV, and a zero prediction MV, which is an MV having a value of zero.

Next, one MV is selected from a plurality of prediction MVs registered in the prediction MV list to determine the MV of the target block.

{Circle around (2)} The variable-length encoding unit describes and encodes a signal “merge_idx”, which is a signal indicating which prediction MV is selected, in a stream.

The prediction MV registered in the prediction MV list described with reference to FIG. 21 is an example, and may be different from the number in the figure, or may not include some types of the prediction MV in the figure, The configuration may be such that a prediction MV other than the type of the prediction MV in the drawing is added.

The final MV may be determined by performing a DMVR (decoder / motion / vector / refinement) process described later using the MV of the target block derived in the merge mode.

Note that the prediction MV candidate is the above-described candidate MV, and the prediction MV list is the above-described candidate MV list. Further, the candidate MV list may be referred to as a candidate list. Further, merge_idx is MV selection information.

[MV derivation> FRUC mode]
The motion information may be derived on the decoding device side without being signalized from the encoding device side. As described above, H.264 A merge mode defined by the H.265 / HEVC standard may be used. Further, for example, the motion information may be derived by performing a motion search on the decoding device side. In the embodiment, on the decoding device side, a motion search is performed without using the pixel values of the current block.

Here, the mode in which the decoding device performs a motion search will be described. The mode in which a motion search is performed on the decoding device side is sometimes referred to as a PMMVD (pattern matched motion vector derivation) mode or a FRUC (frame rate up-conversion) mode.

FIG. 22 shows an example of the FRUC process in the form of a flowchart. First, referring to the motion vector of an encoded block spatially or temporally adjacent to the current block, a list of a plurality of candidates each having a predicted motion vector (MV) (that is, a candidate MV list, (Which may be common with the merge list) is generated (step Si_1). Next, the best candidate MV is selected from a plurality of candidate MVs registered in the candidate MV list (step Si_2). For example, the evaluation value of each candidate MV included in the candidate MV list is calculated, and one candidate MV is selected based on the evaluation value. Then, a motion vector for the current block is derived based on the selected candidate motion vector (step Si_4). Specifically, for example, the motion vector of the selected candidate (best candidate MV) is directly derived as a motion vector for the current block. Further, for example, a motion vector for the current block may be derived by performing pattern matching in a peripheral area of a position in the reference picture corresponding to the selected candidate motion vector. That is, a search using a pattern matching and an evaluation value in a reference picture is performed on a region around the best candidate MV, and if there is an MV having a better evaluation value, the best candidate MV is assigned to the MV. It may be updated and set as the final MV of the current block. It is also possible to adopt a configuration in which processing for updating to an MV having a better evaluation value is not performed.

Finally, the inter prediction unit 126 generates a predicted image of the current block by performing motion compensation on the current block using the derived MV and the encoded reference picture (step Si_5).

場合 When the processing is performed in sub-block units, the same processing may be performed.

The evaluation value may be calculated by various methods. For example, a reconstructed image of a region in a reference picture corresponding to a motion vector and a predetermined region (for example, the region is a region of another reference picture or a region of a block adjacent to the current picture as described below). May be compared with the reconstructed image. The predetermined area may be determined in advance.

Then, the difference between the pixel values of the two reconstructed images may be calculated and used as the evaluation value of the motion vector. The evaluation value may be calculated using other information in addition to the difference value.

Next, an example of pattern matching will be described in detail. First, one candidate MV included in a candidate MV list (for example, a merge list) is selected as a start point of search by pattern matching. For example, the first pattern matching or the second pattern matching may be used as the pattern matching. The first pattern matching and the second pattern matching may be referred to as bilateral matching and template matching, respectively.

[MV derivation>FRUC> bilateral matching]
In the first pattern matching, pattern matching is performed between two blocks in two different reference pictures and along a motion trajectory of the current block (motion trajectory). Therefore, in the first pattern matching, an area in another reference picture along the motion trajectory of the current block is used as a predetermined area for calculating the above-described candidate evaluation value. The predetermined area may be determined in advance.

FIG. 23 is a conceptual diagram for describing an example of first pattern matching (bilateral matching) between two blocks in two reference pictures along a motion trajectory. As shown in FIG. 23, in the first pattern matching, two blocks along the motion trajectory of the current block (Cur @ block) and a pair of two blocks in two different reference pictures (Ref0, Ref1) are used. By searching for the best matching pair, two motion vectors (MV0, MV1) are derived. Specifically, for the current block, a reconstructed image at a specified position in a first encoded reference picture (Ref0) specified by a candidate MV, and a symmetric MV obtained by scaling the candidate MV at a display time interval , A difference from the reconstructed image at the designated position in the second encoded reference picture (Ref1) designated by the above is derived, and an evaluation value is calculated using the obtained difference value. The candidate MV having the best evaluation value among the plurality of candidate MVs can be selected as the final MV, and a good result can be obtained.

Under the assumption of a continuous motion trajectory, the motion vector (MV0, MV1) pointing to two reference blocks is the temporal distance between the current picture (Cur @ Pic) and the two reference pictures (Ref0, Ref1). (TD0, TD1). For example, if the current picture is temporally located between two reference pictures and the temporal distances from the current picture to the two reference pictures are equal, in the first pattern matching, a reflection-symmetric bidirectional motion vector is used. Is derived.

[MV derivation>FRUC> template matching]
In the second pattern matching (template matching), pattern matching is performed between a template in the current picture (a block adjacent to the current block in the current picture (for example, an upper and / or left adjacent block)) and a block in the reference picture. Done. Therefore, in the second pattern matching, a block adjacent to the current block in the current picture is used as a predetermined area for calculating the above-described candidate evaluation value.

FIG. 24 is a conceptual diagram illustrating an example of pattern matching (template matching) between a template in a current picture and a block in a reference picture. As shown in FIG. 24, in the second pattern matching, the current block (Cur @ Pic) is searched for a block that matches the block adjacent to the current block (Cur @ block) in the reference picture (Ref0), thereby searching for the current block. Are derived. Specifically, with respect to the current block, the reconstructed image of the encoded area of the left adjacent area and / or the upper adjacent area and the equivalent image in the encoded reference picture (Ref0) designated by the candidate MV A difference from the reconstructed image at the position is derived, an evaluation value is calculated using the obtained difference value, and the candidate MV having the best evaluation value among the plurality of candidate MVs is selected as the best candidate MV. It is possible.

情報 Information indicating whether or not to apply such a FRUC mode (for example, called a FRUC flag) may be signaled at the CU level. When the FRUC mode is applied (for example, when the FRUC flag is true), information indicating an applicable pattern matching method (first pattern matching or second pattern matching) may be signaled at the CU level. . The signalization of these pieces of information does not need to be limited to the CU level, and may be at another level (for example, a sequence level, a picture level, a slice level, a tile level, a CTU level, or a sub-block level). .

[MV derivation> affine mode]
Next, an affine mode for deriving a motion vector in sub-block units based on motion vectors of a plurality of adjacent blocks will be described. This mode may be referred to as an affine motion compensation prediction mode.

FIG. 25A is a conceptual diagram illustrating an example of deriving a motion vector in sub-block units based on motion vectors of a plurality of adjacent blocks. In FIG. 25A, the current block includes 16 4 × 4 sub-blocks. Here, the motion vector v ₀ of the upper left corner control point of the current block is derived based on the motion vector of the adjacent block, and similarly, the motion vector v ₀ of the upper right corner control point of the current block is calculated based on the motion vector of the adjacent sub block. ₁ is derived. Then, according to the following equation (1A), two motion vectors v ₀ and v ₁ may be projected, and a motion vector (v _x , v _y ) of each sub-block in the current block may be derived.

Here, x and y indicate the horizontal position and the vertical position of the sub-block, respectively, and w indicates a predetermined weighting factor. The predetermined weight coefficient may be determined in advance.

情報 Information indicating such an affine mode (for example, called an affine flag) may be signaled at the CU level. The signaling of the information indicating the affine mode need not be limited to the CU level, but may be at another level (for example, a sequence level, a picture level, a slice level, a tile level, a CTU level, or a sub-block level). You may.

In addition, such an affine mode may include several modes in which the method of deriving the motion vector of the upper left and upper right corner control points is different. For example, the affine mode includes two modes: an affine inter (also called an affine normal inter) mode and an affine merge mode.

[MV derivation> affine mode]
FIG. 25B is a conceptual diagram for explaining an example of deriving a motion vector in subblock units in the affine mode having three control points. In FIG. 25B, the current block includes 16 4 × 4 sub-blocks. Here, the motion vector v ₀ of the upper left corner control point of the current block is derived based on the motion vector of the neighboring block, and similarly, the motion vector v ₁ of the upper right corner control point of the current block is derived based on the motion vector of the neighboring block. , motion vector v ₂ in the lower left angle control point in the current block based on the motion vector of the neighboring block is derived. Then, according to the following equation (1B), three motion vectors v ₀ , v ₁ and v ₂ may be projected, and the motion vector (v _x , v _y ) of each sub-block in the current block is derived. Is also good.

Here, x and y indicate the horizontal position and the vertical position of the center of the sub-block, respectively, w indicates the width of the current block, and h indicates the height of the current block.

Affine modes with different numbers of control points (eg, two and three) may be signaled by switching at the CU level. The information indicating the number of control points in the affine mode used at the CU level may be signaled at another level (for example, a sequence level, a picture level, a slice level, a tile level, a CTU level, or a sub-block level). Good.

The affine mode having three control points may include some modes in which the method of deriving the motion vectors of the upper left, upper right, and lower left corner control points is different. For example, the affine mode includes two modes: an affine inter (also called an affine normal inter) mode and an affine merge mode.

[MV derivation> affine merge mode]
26A, 26B, and 26C are conceptual diagrams for explaining the affine merge mode.

In the affine merge mode, as shown in FIG. 26A, for example, encoded block A (left), block B (upper), block C (upper right), block D (lower left), and block E (upper left) adjacent to the current block ), The respective predicted motion vectors of the control points of the current block are calculated based on a plurality of motion vectors corresponding to the block encoded in the affine mode. Specifically, these blocks are checked in the order of coded block A (left), block B (upper), block C (upper right), block D (lower left), and block E (upper left), and in affine mode The first valid block encoded is identified. A predicted motion vector of the control point of the current block is calculated based on the plurality of motion vectors corresponding to the specified block.

For example, as shown in FIG. 26B, when a block A adjacent to the left of the current block is encoded in the affine mode having two control points, the upper left corner and the upper right corner of the encoded block including the block A motion projected onto the position vector v ₃ and v ₄ is derived. Then, the motion vector v ₃ and v ₄ derived, the predicted motion vector v ₀ of the control point of the upper left corner of the current block, the prediction motion vector v ₁ of the control point in the upper right corner is calculated.

For example, as shown in FIG. 26C, when the block A adjacent to the left of the current block is encoded in the affine mode having three control points, the upper left corner and the upper right corner of the encoded block including the block A And the motion vectors v ₃ , v ₄ and v ₅ projected at the position of the lower left corner. Then, from the derived motion vectors v ₃ , v ₄ and v ₅ , the predicted motion vector v ₀ of the control point at the upper left corner of the current block, the predicted motion vector v ₁ of the control point at the upper right corner, and the control of the lower left corner are calculated. predicted motion vector v ₂ of the points are calculated.

This prediction motion vector derivation method may be used to derive a prediction motion vector for each control point of the current block in step Sj_1 in FIG. 29 described below.

FIG. 27 is a flowchart showing an example of the affine merge mode.

In the フィン affine merge mode, as illustrated, first, the inter prediction unit 126 derives each prediction MV of the control point of the current block (step Sk_1). The control points are points at the upper left and upper right corners of the current block as shown in FIG. 25A, or points at the upper left, upper right and lower left corners of the current block as shown in FIG. 25B.

That is, as illustrated in FIG. 26A, the inter prediction unit 126 performs the order of the coded block A (left), block B (upper), block C (upper right), block D (lower left), and block E (upper left). Examine these blocks and identify the first valid block encoded in affine mode.

Then, when the block A is specified and the block A has two control points, as illustrated in FIG. 26B, the inter prediction unit 126 calculates the motion vector v _{3 of the} upper left corner and the upper right corner of the encoded block including the block A. and v _4, and calculates a motion vector v ₀ of the control point of the upper left corner of the current block, the control point in the upper right corner and a motion vector v _1. For example, the inter prediction unit 126 projects the motion vectors v ₃ and v ₄ at the upper left corner and the upper right corner of the coded block onto the current block, and thereby the predicted motion vector v _{0 at} the control point at the upper left corner of the current block. If, to calculate the predicted motion vector v ₁ of the control point in the upper right corner.

Alternatively, when the block A is specified and the block A has three control points, as illustrated in FIG. 26C, the inter prediction unit 126 performs the motion of the upper left corner, the upper right corner, and the lower left corner of the encoded block including the block A. From the vectors v ₃ , v ₄ and v ₅ , the motion vector v ₀ of the control point at the upper left corner of the current block, the motion vector v ₁ of the control point at the upper right corner, and the motion vector v ₂ of the control point at the lower left corner are calculated. I do. For example, the inter prediction unit 126 projects the motion vectors v ₃ , v _4, and v ₅ of the upper left corner, the upper right corner, and the lower left corner of the encoded block onto the current block, thereby controlling the control point of the upper left corner of the current block. to the calculated and the predicted motion vector v _0, the predicted motion vector v ₁ of the control point in the upper right _corner, the control point of the lower-left corner of the motion vector v _2.

Next, the inter prediction unit 126 performs motion compensation on each of the plurality of sub-blocks included in the current block. That is, the inter prediction unit 126 calculates, for each of the plurality of sub-blocks, two predicted motion vectors v ₀ and v ₁ and the above equation (1A) or three predicted motion vectors v ₀ , v ₁ and v ₂ . Using the above equation (1B), the motion vector of the sub-block is calculated as the affine MV (step Sk_2). Then, the inter prediction unit 126 performs motion compensation on the sub-block using the affine MV and the encoded reference picture (step Sk_3). As a result, motion compensation is performed on the current block, and a predicted image of the current block is generated.

[MV Derivation> Affine Inter Mode]
FIG. 28A is a conceptual diagram for describing an affine inter mode having two control points.

In the affine inter mode, as shown in FIG. 28A, a motion vector selected from the motion vectors of coded blocks A, B, and C adjacent to the current block is used to predict the control point at the upper left corner of the current block. It is used as the motion vector _{v 0.} Similarly, motion vectors selected from the motion vectors of the encoded block D and block E is adjacent to the current block are used as predicted motion vector v ₁ of the control point of the upper-right corner of the current block.

FIG. 28B is a conceptual diagram illustrating an affine inter mode having three control points.

In this affine inter mode, as shown in FIG. 28B, a motion vector selected from the motion vectors of the coded blocks A, B and C adjacent to the current block is used to predict the control point at the upper left corner of the current block. It is used as the motion vector _{v 0.} Similarly, motion vectors selected from the motion vectors of the encoded block D and block E is adjacent to the current block are used as predicted motion vector v ₁ of the control point of the upper-right corner of the current block. Moreover, motion vectors selected from the motion vectors of the encoded block F and block G adjacent to the current block are used as predicted motion vector v ₂ of the control points of the lower left corner of the current block.

FIG. 29 is a flowchart showing an example of the affine inter mode.

As illustrated, in the affine inter mode, first, the inter prediction unit 126 calculates the prediction MV (v ₀ , v ₁ ) or (v ₀ , v ₁ , v) of each of two or three control points of the current block. ₂ ) is derived (step Sj_1). The control point is a point at the upper left corner, upper right corner or lower left corner of the current block as shown in FIG. 25A or 25B.

That is, the inter prediction unit 126 selects the motion vector of one of the encoded blocks near each control point of the current block shown in FIG. 28A or FIG. 28B, thereby predicting the control point of the current block. The motion vector (v ₀ , v ₁ ) or (v ₀ , v ₁ , v ₂ ) is derived. At this time, the inter prediction unit 126 encodes predicted motion vector selection information for identifying the two selected motion vectors into a stream.

For example, the inter prediction unit 126 determines which motion vector of the encoded block adjacent to the current block is to be selected as the predicted motion vector of the control point by using a cost evaluation or the like, and determines which predicted motion vector A flag indicating the selection may be described in the bit stream.

Next, the inter prediction unit 126 performs a motion search (steps Sj_3 and Sj_4) while updating each of the predicted motion vectors selected or derived in step Sj_1 (step Sj_2). That is, the inter prediction unit 126 calculates the motion vector of each sub-block corresponding to the predicted motion vector to be updated as the affine MV using the above equation (1A) or equation (1B) (step Sj_3). Then, the inter prediction unit 126 performs motion compensation on each sub-block using the affine MV and the coded reference picture (step Sj_4). As a result, in the motion search loop, the inter prediction unit 126 determines, for example, a predicted motion vector at which the lowest cost is obtained as the control point motion vector (step Sj_5). At this time, the inter prediction unit 126 further encodes a difference value between the determined MV and the predicted motion vector into a stream as a difference MV.

Finally, the inter prediction unit 126 generates a predicted image of the current block by performing motion compensation on the current block using the determined MV and the encoded reference picture (step Sj_6).

[MV Derivation> Affine Inter Mode]
In the case where affine modes with different numbers of control points (for example, two and three) are switched at the CU level to signal, the number of control points may differ between the coded block and the current block. FIG. 30A and FIG. 30B are conceptual diagrams for explaining a method of deriving a predicted vector of a control point when the number of control points differs between an encoded block and a current block.

For example, as shown in FIG. 30A, the current block has three control points of an upper left corner, an upper right corner, and a lower left corner, and a block A adjacent to the left of the current block is encoded in an affine mode having two control points. If it is, the motion vector v ₃ and v ₄ projected onto the position of the upper left corner and upper right corner of the encoded blocks containing the block a is derived. Then, the motion vector v ₃ and v ₄ derived, the predicted motion vector v ₀ of the control point of the upper left corner of the current block, the prediction motion vector v ₁ of the control point in the upper right corner is calculated. Furthermore, the motion vector v ₀ and v ₁ derived, predicted motion vector v ₂ of the control point of the bottom left corner is calculated.

For example, as shown in FIG. 30B, the current block has two control points of an upper left corner and an upper right corner, and a block A adjacent to the left of the current block is encoded in an affine mode having three control points. In this case, motion vectors v ₃ , v ₄ and v ₅ projected at the upper left corner, upper right corner and lower left corner of the encoded block including block A are derived. Then, from the derived motion vectors v ₃ , v ₄ and v ₅ , a predicted motion vector v ₀ of the control point at the upper left corner of the current block and a predicted motion vector v ₁ of the control point at the upper right corner of the current block are calculated.

This prediction motion vector derivation method may be used to derive each prediction motion vector of the control point of the current block in step Sj_1 in FIG.

[MV derivation> DMVR]
FIG. 31A is a flowchart showing the relationship between the merge mode and DMVR.

The inter prediction unit 126 derives a motion vector of the current block in the merge mode (Step Sl_1). Next, the inter prediction unit 126 determines whether or not to search for a motion vector, that is, whether to perform a motion search (step Sl_2). Here, when the inter prediction unit 126 determines that the motion search is not performed (No in Step Sl_2), the inter prediction unit 126 determines the motion vector derived in Step Sl_1 as the final motion vector for the current block (Step Sl_4). That is, in this case, the motion vector of the current block is determined in the merge mode.

On the other hand, if it is determined in step Sl_1 that a motion search is to be performed (Yes in step Sl_2), the inter prediction unit 126 searches for a peripheral region of the reference picture indicated by the motion vector derived in step Sl_1, thereby obtaining the current block. On the other hand, a final motion vector is derived (step Sl_3). That is, in this case, the motion vector of the current block is determined by the DMVR.

FIG. 31B is a conceptual diagram for explaining an example of the DMVR process for determining the MV.

First, the optimal MVP set in the current block (for example, in the merge mode) is set as a candidate MV. Then, a reference pixel is specified from the first reference picture (L0), which is a coded picture in the L0 direction, according to the candidate MV (L0). Similarly, a reference pixel is specified from the second reference picture (L1), which is a coded picture in the L1 direction, according to the candidate MV (L1). A template is generated by averaging these reference pixels.

{Circle around (2)} Next, using the template, the peripheral areas of the candidate MVs of the first reference picture (L0) and the second reference picture (L1) are respectively searched, and the MV having the minimum cost is determined as the final MV. The cost value may be calculated using, for example, a difference value between each pixel value of the template and each pixel value of the search area, a candidate MV value, and the like.

Note that typically, the configuration and operation of the processing described here are basically common between an encoding device and a decoding device described later.

ても Any processing may be used as long as it can search the vicinity of the candidate MV and derive the final MV without being the processing example itself described here.

[Motion Compensation> BIO / OBMC]
In the motion compensation, there is a mode for generating a predicted image and correcting the predicted image. The modes are, for example, BIO and OBMC described later.

FIG. 32 is a flowchart illustrating an example of generation of a predicted image.

The inter prediction unit 126 generates a predicted image (Step Sm_1), and corrects the predicted image by, for example, one of the above-described modes (Step Sm_2).

FIG. 33 is a flowchart showing another example of generation of a predicted image.

The inter prediction unit 126 determines the motion vector of the current block (Step Sn_1). Next, the inter prediction unit 126 generates a predicted image (Step Sn_2), and determines whether or not to perform a correction process (Step Sn_3). Here, when the inter prediction unit 126 determines that the correction process is to be performed (Yes in Step Sn_3), the inter prediction unit 126 corrects the predicted image to generate a final predicted image (Step Sn_4). On the other hand, when determining that the correction process is not performed (No in Step Sn_3), the inter prediction unit 126 outputs the predicted image as a final predicted image without correction (Step Sn_5).

動き In motion compensation, there is a mode for correcting the luminance when generating a predicted image. The mode is, for example, LIC described later.

FIG. 34 is a flowchart showing another example of generating a predicted image.

The inter prediction unit 126 derives a motion vector of the current block (Step So_1). Next, the inter prediction unit 126 determines whether to perform the luminance correction process (Step So_2). Here, when determining that the luminance correction process is to be performed (Yes in Step So_2), the inter prediction unit 126 generates a predicted image while performing the luminance correction (Step So_3). That is, a predicted image is generated by the LIC. On the other hand, when determining that the luminance correction process is not to be performed (No in Step So_2), the inter prediction unit 126 generates a predicted image by normal motion compensation without performing luminance correction (Step So_4).

[Motion compensation> OBMC]
The inter prediction signal may be generated using not only the motion information of the current block obtained by the motion search but also the motion information of the adjacent block. Specifically, by weighting and adding a prediction signal based on motion information obtained by motion search (within a reference picture) and a prediction signal based on motion information of an adjacent block (within a current picture), An inter prediction signal may be generated for each sub-block in a block. Such inter prediction (motion compensation) may be called OBMC (overlapped block motion compensation).

In the OBMC mode, information indicating the size of a sub-block for OBMC (for example, called OBMC block size) may be signaled at a sequence level. Further, information indicating whether to apply the OBMC mode (for example, called an OBMC flag) may be signaled at the CU level. The level of signalization of these pieces of information need not be limited to the sequence level and the CU level, but may be another level (eg, picture level, slice level, tile level, CTU level, or sub-block level). Good.

An example of the OBMC mode will be described more specifically. FIG. 35 and FIG. 36 are a flowchart and a conceptual diagram for explaining the outline of the predicted image correction processing by the OBMC processing.

First, as shown in FIG. 36, a predicted image (Pred) by normal motion compensation is obtained using a motion vector (MV) assigned to a processing target (current) block. In FIG. 36, an arrow “MV” indicates a reference picture, and indicates what the current block of the current picture refers to to obtain a predicted image.

Next, the motion vector (MV_L) already derived for the encoded left adjacent block is applied (reused) to the current block to obtain a predicted image (Pred_L). The motion vector (MV_L) is indicated by an arrow “MV_L” pointing from the current block to a reference picture. Then, the first correction of the predicted image is performed by overlapping the two predicted images Pred and Pred_L. This has the effect of mixing the boundaries between adjacent blocks.

Similarly, the motion vector (MV_U) already derived for the encoded upper adjacent block is applied (reused) to the current block to obtain a predicted image (Pred_U). The motion vector (MV_U) is indicated by an arrow “MV_U” pointing from the current block to a reference picture. Then, a second correction of the predicted image is performed by superimposing the predicted image Pred_U on the predicted image (for example, Pred and Pred_L) on which the first correction has been performed. This has the effect of mixing the boundaries between adjacent blocks. The predicted image obtained by the second correction is the final predicted image of the current block in which the boundary with the adjacent block has been mixed (smoothed).

Note that the above example is a two-pass correction method using left-adjacent and upper-adjacent blocks, but the correction method is three-pass or more paths using right-adjacent and / or lower-adjacent blocks. May be used.

Note that the region to be superimposed may not be the pixel region of the entire block, but may be only a partial region near the block boundary.

Here, the prediction image correction processing of the OBMC for obtaining one prediction image Pred by superimposing additional prediction images Pred_L and Pred_U from one reference picture has been described. However, when a predicted image is corrected based on a plurality of reference images, a similar process may be applied to each of the plurality of reference pictures. In such a case, by performing OBMC image correction based on a plurality of reference pictures, a corrected prediction image is obtained from each reference picture, and then the obtained plurality of corrected prediction images are further superimposed. To obtain the final predicted image.

In the OBMC, the unit of the target block may be a prediction block unit or a sub-block unit obtained by further dividing the prediction block.

As a method of determining whether to apply the OBMC process, for example, there is a method of using obmc_flag, which is a signal indicating whether to apply the OBMC process. As a specific example, the encoding device may determine whether the target block belongs to a region having a complicated motion. The encoding apparatus sets the value 1 as obmc_flag to perform encoding by applying the OBMC process when belonging to a complicated motion region, and performs obmc_flag when not belonging to a complicated motion region. By setting the value to 0, the block is encoded without applying the OBMC process. On the other hand, the decoding device decodes obmc_flag described in a stream (for example, a compressed sequence), and performs decoding by switching whether or not to apply the OBMC process according to the value.

In the example described above, the inter prediction unit 126 generates one rectangular predicted image for the rectangular current block. However, the inter prediction unit 126 generates a plurality of predicted images having a shape different from the rectangle for the rectangular current block, and generates a final rectangular predicted image by combining the plurality of predicted images. May be. The shape different from the rectangle may be, for example, a triangle.

FIG. 37 is a conceptual diagram for describing generation of two triangular predicted images.

The inter prediction unit 126 generates a predicted image of a triangle by performing motion compensation on the first partition of the triangle in the current block using the first MV of the first partition. Similarly, the inter prediction unit 126 generates a triangle predicted image by performing motion compensation on the second partition of the triangle in the current block using the second MV of the second partition. Then, the inter prediction unit 126 combines these prediction images to generate a prediction image having the same rectangle as the current block.

In the example shown in FIG. 37, the first partition and the second partition are each triangular, but may be trapezoidal or different from each other. Further, in the example shown in FIG. 37, the current block is composed of two partitions, but may be composed of three or more partitions.

{Circle around (1)} The first partition and the second partition may overlap. That is, the first partition and the second partition may include the same pixel region. In this case, a predicted image of the current block may be generated using the predicted image in the first partition and the predicted image in the second partition.

{Although this example shows an example in which a predicted image is generated by inter prediction for both partitions, a predicted image may be generated by intra prediction for at least one partition.

[Motion Compensation> BIO]
Next, a method for deriving a motion vector will be described. First, a mode for deriving a motion vector based on a model assuming uniform linear motion will be described. This mode may be referred to as a BIO (bi-directional optical flow) mode.

FIG. 38 is a conceptual diagram for describing a model assuming constant velocity linear motion. In FIG. 38, (vx, vy) indicates a velocity vector, and τ0 and τ1 indicate temporal distances between the current picture (Cur @ Pic) and two reference pictures (Ref0, Ref1). (MVx0, MVy0) indicates a motion vector corresponding to the reference picture Ref0, and (MVx1, MVy1) indicates a motion vector corresponding to the reference picture Ref1.

At this time, under the assumption of constant velocity linear motion of the velocity vector (vx, vy), (MVx0, MVy0) and (MVx1, MVy1) are (vxτ0, vyτ0) and (−vxτ1, −vyτ1), respectively. Then, the following optical flow equation (2) may be adopted.

Here, I (k) indicates the luminance value of the reference image k (k = 0, 1) after motion compensation. This optical flow equation includes (i) the time derivative of the luminance value, (ii) the product of the horizontal velocity and the horizontal component of the spatial gradient of the reference image, and (iii) the vertical velocity and the spatial gradient of the reference image. This shows that the sum of the product of the vertical components of and is equal to zero. Based on a combination of the optical flow equation and Hermite interpolation, a block-by-block motion vector obtained from a merge list or the like may be corrected in pixel units.

Note that the motion vector may be derived on the decoding device side by a method different from the method for deriving the motion vector based on a model assuming uniform linear motion. For example, a motion vector may be derived for each sub-block based on motion vectors of a plurality of adjacent blocks.

[Motion compensation> LIC]
Next, an example of a mode for generating a prediction image (prediction) using LIC (local illumination compensation) processing will be described.

FIG. 39 is a conceptual diagram illustrating an example of a predicted image generation method using a luminance correction process by an LIC process.

First, the MV is derived from the encoded reference picture, and the reference image corresponding to the current block is obtained.

Next, for the current block, information indicating how the luminance value has changed between the reference picture and the current picture is extracted. This extraction is performed by extracting the luminance pixel values of the encoded left adjacent reference area (peripheral reference area) and the encoded upper adjacent reference area (peripheral reference area) of the current picture, and the luminance value of the reference picture specified by the derived MV. This is performed based on the luminance pixel value at the equivalent position. Then, a luminance correction parameter is calculated using information indicating how the luminance value has changed.

A predicted image for the current block is generated by performing a luminance correction process that applies the luminance correction parameter to a reference image in a reference picture specified by $ MV.

Note that the shape of the peripheral reference area in FIG. 39 is an example, and other shapes may be used.

Also, here, the process of generating a predicted image from one reference picture has been described. However, the same applies to the case of generating a predicted image from a plurality of reference pictures. The predicted image may be generated after performing the luminance correction processing in the same manner as described above.

As a method of determining whether or not to apply the LIC processing, for example, there is a method of using lic_flag which is a signal indicating whether or not to apply the LIC processing. As a specific example, in the encoding device, it is determined whether the current block belongs to a region where a luminance change occurs. If the current block belongs to a region where a luminance change occurs, a value is set as lic_flag. The coding is performed by setting 1 and applying the LIC processing, and when the pixel does not belong to the area where the luminance change occurs, the value 0 is set as lic_flag and the coding is performed without applying the LIC processing. On the other hand, the decoding device may decode the lic_flag described in the stream, and perform decoding by switching whether or not to apply the LIC processing according to the value.

As another method of determining whether or not to apply the LIC processing, for example, there is a method of determining according to whether or not the LIC processing is applied to a peripheral block. As a specific example, when the current block is in the merge mode, it is determined whether or not the peripheral encoded block selected at the time of derivation of the MV in the merge mode processing is encoded by applying the LIC processing. . The coding is performed by switching whether or not to apply the LIC processing according to the result. In this case, the same processing is applied to the processing on the decoding device side.

The mode of the LIC processing (luminance correction processing) has been described with reference to FIG. 39, and the details thereof will be described below.

First, the inter prediction unit 126 derives a motion vector for acquiring a reference image corresponding to the current block from a reference picture that is a coded picture.

Next, the inter prediction unit 126 calculates the luminance pixel values of the encoded neighboring reference regions on the left and upper sides of the current block and the luminance pixels at the same position in the reference picture specified by the motion vector. Using the value, information indicating how the luminance value has changed between the reference picture and the current picture is extracted to calculate a luminance correction parameter. For example, the luminance pixel value of a certain pixel in the peripheral reference area in the encoding target picture is p0, and the luminance pixel value of a pixel in the peripheral reference area in the reference picture at the same position as the pixel is p1. The inter prediction unit 126 calculates coefficients A and B for optimizing A × p1 + B = p0 as a luminance correction parameter for a plurality of pixels in the peripheral reference area.

Next, the inter prediction unit 126 performs a luminance correction process on the reference image in the reference picture specified by the motion vector using the luminance correction parameter, thereby generating a predicted image for the encoding target block. For example, the luminance pixel value in the reference image is p2, and the luminance pixel value of the predicted image after the luminance correction processing is p3. The inter prediction unit 126 generates a predicted image after the luminance correction processing by calculating A × p2 + B = p3 for each pixel in the reference image.

Note that the shape of the peripheral reference area in FIG. 39 is an example, and other shapes may be used. A part of the peripheral reference area shown in FIG. 39 may be used. For example, an area including a predetermined number of pixels thinned out from each of the upper adjacent pixel and the left adjacent pixel may be used as the peripheral reference area. Further, the peripheral reference area is not limited to the area adjacent to the encoding target block, and may be an area not adjacent to the encoding target block. The predetermined number of pixels may be determined in advance.

In the example shown in FIG. 39, the peripheral reference area in the reference picture is an area specified by the motion vector of the current picture from the peripheral reference area in the current picture. It may be a designated area. For example, the other motion vector may be a motion vector of a peripheral reference area in the current picture.

Here, the operation in the encoding device 100 has been described, but the operation in the decoding device 200 is also typically the same.

Note that the LIC processing may be applied to not only the luminance but also the color difference. At this time, a correction parameter may be derived individually for each of Y, Cb, and Cr, or a common correction parameter may be used for any of them.

Ｌ Also, the LIC processing may be applied on a sub-block basis. For example, the correction parameter may be derived using the peripheral reference area of the current sub-block and the peripheral reference area of the reference sub-block in the reference picture specified by the MV of the current sub-block.

[Prediction control unit]
The prediction control unit 128 selects one of an intra prediction signal (a signal output from the intra prediction unit 124) and an inter prediction signal (a signal output from the inter prediction unit 126), and subtracts the selected signal as a prediction signal. Output to the section 104 and the addition section 116.

As shown in FIG. 1, in various coding apparatuses, the prediction control unit 128 may output a prediction parameter input to the entropy coding unit 110. The entropy coding unit 110 may generate a coded bit stream (or sequence) based on the prediction parameters input from the prediction control unit 128 and the quantization coefficients input from the quantization unit 108. The prediction parameter may be used for a decoding device. The decoding device may receive and decode the encoded bit stream, and perform the same processing as the prediction processing performed in the intra prediction unit 124, the inter prediction unit 126, and the prediction control unit 128. The prediction parameter is a selected prediction signal (for example, a motion vector, a prediction type, or a prediction mode used in the intra prediction unit 124 or the inter prediction unit 126), or the intra prediction unit 124, the inter prediction unit 126, and the prediction control unit. Any index, flag, or value based on or indicative of the prediction process performed at 128 may be included.

[Implementation example of encoding device]
FIG. 40 is a block diagram illustrating an implementation example of the encoding device 100. The encoding device 100 includes a processor a1 and a memory a2. For example, a plurality of components of the encoding device 100 shown in FIG. 1 are implemented by the processor a1 and the memory a2 shown in FIG.

The processor a1 is a circuit that performs information processing, and is a circuit that can access the memory a2. For example, the processor a1 is a dedicated or general-purpose electronic circuit that encodes a moving image. The processor a1 may be a processor such as a CPU. Further, the processor a1 may be an aggregate of a plurality of electronic circuits. Further, for example, the processor a1 may play the role of a plurality of components among the plurality of components of the encoding device 100 illustrated in FIG. 1 and the like.

The memory a2 is a dedicated or general-purpose memory in which information for the processor a1 to encode a moving image is stored. The memory a2 may be an electronic circuit, and may be connected to the processor a1. Further, the memory a2 may be included in the processor a1. Further, the memory a2 may be an aggregate of a plurality of electronic circuits. The memory a2 may be a magnetic disk, an optical disk, or the like, or may be expressed as a storage or a recording medium. Further, the memory a2 may be a nonvolatile memory or a volatile memory.

For example, the memory a2 may store a moving image to be coded, or may store a bit string corresponding to the coded moving image. Further, the memory a2 may store a program for the processor a1 to encode a moving image.

For example, the memory a2 may serve as a component for storing information among a plurality of components of the encoding device 100 illustrated in FIG. 1 and the like. For example, the memory a2 may serve as the block memory 118 and the frame memory 122 shown in FIG. More specifically, the memory a2 may store reconstructed blocks, reconstructed pictures, and the like.

In the encoding device 100, not all of the plurality of components illustrated in FIG. 1 and the like need to be implemented, and all of the plurality of processes described above need not be performed. Some of the components illustrated in FIG. 1 and the like may be included in another device, or some of the above-described processes may be performed by another device.

[Decryption device]
Next, a decoding device capable of decoding an encoded signal (encoded bit stream) output from the encoding device 100 will be described. FIG. 41 is a block diagram illustrating a functional configuration of the decoding device 200 according to the embodiment. The decoding device 200 is a moving image decoding device that decodes a moving image in block units.

As shown in FIG. 41, the decoding device 200 includes an entropy decoding unit 202, an inverse quantization unit 204, an inverse transformation unit 206, an addition unit 208, a block memory 210, a loop filter unit 212, and a frame memory 214. , An intra prediction unit 216, an inter prediction unit 218, and a prediction control unit 220.

The decoding device 200 is realized by, for example, a general-purpose processor and a memory. In this case, when the software program stored in the memory is executed by the processor, the processor includes an entropy decoding unit 202, an inverse quantization unit 204, an inverse transformation unit 206, an addition unit 208, a loop filter unit 212, an intra prediction unit 216, and functions as the inter prediction unit 218 and the prediction control unit 220. Further, the decoding device 200 is a dedicated device corresponding to the entropy decoding unit 202, the inverse quantization unit 204, the inverse transform unit 206, the addition unit 208, the loop filter unit 212, the intra prediction unit 216, the inter prediction unit 218, and the prediction control unit 220. May be realized as one or more electronic circuits.

Hereinafter, after describing the overall processing flow of the decoding device 200, each component included in the decoding device 200 will be described.

[Overall flow of decryption process]
FIG. 42 is a flowchart illustrating an example of the entire decoding process performed by the decoding device 200.

First, the entropy decoding unit 202 of the decoding device 200 specifies a division pattern of a fixed-size block (for example, 128 × 128 pixels) (Step Sp_1). This division pattern is the division pattern selected by the encoding device 100. Then, the decoding device 200 performs the processing of steps Sp_2 to Sp_6 on each of the plurality of blocks constituting the divided pattern.

{That is, the entropy decoding unit 202 decodes (specifically, entropy-decodes) the encoded quantization coefficient and the prediction parameter of the decoding target block (also referred to as a current block) (Step Sp_2).

Next, the inverse quantization unit 204 and the inverse transform unit 206 restore the plurality of prediction residuals (that is, difference blocks) by performing inverse quantization and inverse transform on the plurality of quantized coefficients (Step Sp_3). ).

Next, the prediction processing unit including all or a part of the intra prediction unit 216, the inter prediction unit 218, and the prediction control unit 220 generates a prediction signal (also referred to as a prediction block) of the current block (Step Sp_4).

Next, the adding unit 208 reconstructs the current block into a reconstructed image (also referred to as a decoded image block) by adding the prediction block to the difference block (Step Sp_5).

{Circle around (2)} When the reconstructed image is generated, the loop filter unit 212 performs filtering on the reconstructed image (Step Sp_6).

Then, the decoding device 200 determines whether or not decoding of the entire picture has been completed (Step Sp_7), and when determining that the decoding has not been completed (No in Step Sp_7), repeats the processing from Step Sp_1.

よう As shown, the processing of steps Sp_1 to Sp_7 is performed sequentially by the decoding device 200. Alternatively, some of these processes may be performed in parallel, or the order may be changed.

[Entropy decoding unit]
The entropy decoding unit 202 performs entropy decoding on the encoded bit stream. Specifically, for example, the entropy decoding unit 202 arithmetically decodes an encoded bit stream into a binary signal. Then, the entropy decoding unit 202 debinarizes the binary signal. The entropy decoding unit 202 outputs the quantized coefficients to the inverse quantization unit 204 in block units. The entropy decoding unit 202 may output the prediction parameters included in the encoded bit stream (see FIG. 1) to the intra prediction unit 216, the inter prediction unit 218, and the prediction control unit 220 according to the embodiment. The intra prediction unit 216, the inter prediction unit 218, and the prediction control unit 220 can execute the same prediction processing as the processing performed by the intra prediction unit 124, the inter prediction unit 126, and the prediction control unit 128 on the encoding device side.

[Inverse quantization unit]
The inverse quantization unit 204 inversely quantizes a quantization coefficient of a decoding target block (hereinafter, referred to as a current block), which is an input from the entropy decoding unit 202. Specifically, for each of the quantization coefficients of the current block, the inverse quantization unit 204 inversely quantizes the quantization coefficient based on a quantization parameter corresponding to the quantization coefficient. Then, the inverse quantization unit 204 outputs the inversely quantized coefficients (that is, transform coefficients) of the current block to the inverse transform unit 206.

[Inverse conversion unit]
The inverse transform unit 206 restores a prediction error by inversely transforming the transform coefficient input from the inverse quantization unit 204.

For example, if the information read from the coded bit stream indicates that EMT or AMT is applied (for example, the AMT flag is true), the inverse transform unit 206 determines the current block based on the information indicating the read conversion type. Is inversely transformed.

{In addition, for example, when the information read from the encoded bit stream indicates that the NSST is to be applied, the inverse transform unit 206 applies the inverse retransform to the transform coefficient.

[Adder]
The addition unit 208 reconstructs the current block by adding the prediction error input from the inverse conversion unit 206 and the prediction sample input from the prediction control unit 220. Then, the adding unit 208 outputs the reconstructed block to the block memory 210 and the loop filter unit 212.

[Block memory]
The block memory 210 is a storage unit for storing blocks that are referred to in intra prediction and are in a current picture to be decoded (hereinafter, referred to as a current picture). Specifically, the block memory 210 stores the reconstructed block output from the adder 208.

[Loop filter section]
The loop filter unit 212 performs a loop filter on the block reconstructed by the adding unit 208, and outputs the filtered reconstructed block to the frame memory 214, the display device, and the like.

If the ALF on / off information read from the coded bitstream indicates ALF on, one filter is selected from the plurality of filters based on the local gradient direction and activity. The selected filter is applied to the reconstruction block.

[Frame memory]
The frame memory 214 is a storage unit for storing reference pictures used for inter prediction, and may be called a frame buffer. Specifically, the frame memory 214 stores the reconstructed blocks filtered by the loop filter unit 212.

[Prediction processing unit (intra prediction unit / inter prediction unit / prediction control unit)]
FIG. 43 is a flowchart illustrating an example of processing performed by the prediction processing unit of the decoding device 200. The prediction processing unit includes all or some components of the intra prediction unit 216, the inter prediction unit 218, and the prediction control unit 220.

The prediction processing unit generates a predicted image of the current block (Step Sq_1). This prediction image is also called a prediction signal or a prediction block. The prediction signal includes, for example, an intra prediction signal or an inter prediction signal. Specifically, the prediction processing unit generates a reconstructed image that has already been obtained by performing generation of a prediction block, generation of a difference block, generation of a coefficient block, restoration of a difference block, and generation of a decoded image block. To generate a predicted image of the current block.

The reconstructed image may be, for example, an image of a reference picture or an image of a decoded block in the current picture which is a picture including the current block. The decoded block in the current picture is, for example, a block adjacent to the current block.

FIG. 44 is a flowchart illustrating another example of the processing performed by the prediction processing unit of the decoding device 200.

The prediction processing unit determines a method or a mode for generating a predicted image (Step Sr_1). For example, this scheme or mode may be determined based on, for example, a prediction parameter or the like.

When the prediction processing unit determines the first method as the mode for generating the predicted image, the prediction processing unit generates the predicted image according to the first method (Step Sr_2a). Further, when the prediction processing unit determines the second method as the mode for generating the predicted image, the prediction processing unit generates the predicted image according to the second method (Step Sr_2b). In addition, when the prediction processing unit determines the third method as the mode for generating a predicted image, the prediction processing unit generates a predicted image according to the third method (Step Sr_2c).

The first scheme, the second scheme, and the third scheme are different schemes for generating a predicted image, and are, for example, inter prediction schemes, intra prediction schemes, and other prediction schemes, respectively. There may be. In these prediction methods, the above-described reconstructed image may be used.

[Intra prediction unit]
The intra prediction unit 216 performs intra prediction with reference to a block in the current picture stored in the block memory 210 based on the intra prediction mode read from the coded bit stream, thereby obtaining a prediction signal (intra prediction mode). Signal). Specifically, the intra prediction unit 216 generates an intra prediction signal by performing intra prediction with reference to samples (for example, a luminance value and a color difference value) of a block adjacent to the current block, and performs prediction control on the intra prediction signal. Output to the unit 220.

When the intra prediction mode that refers to the luminance block is selected in the intra prediction of the chrominance block, the intra prediction unit 216 may predict the chrominance component of the current block based on the luminance component of the current block. .

In addition, when the information read from the coded bit stream indicates the application of PDPC, the intra prediction unit 216 corrects the pixel value after intra prediction based on the gradient of the reference pixel in the horizontal / vertical directions.

[Inter prediction unit]
The inter prediction unit 218 predicts the current block with reference to the reference picture stored in the frame memory 214. The prediction is performed in units of the current block or sub-blocks (for example, 4 × 4 blocks) in the current block. For example, the inter prediction unit 218 performs motion compensation using motion information (for example, a motion vector) read from a coded bit stream (for example, a prediction parameter output from the entropy decoding unit 202), thereby performing the current block or It generates an inter prediction signal of the sub-block and outputs the inter prediction signal to the prediction control unit 220.

When the information read from the coded bit stream indicates that the OBMC mode is applied, the inter prediction unit 218 uses the motion information of the adjacent block as well as the motion information of the current block obtained by the motion search. , Generate an inter prediction signal.

Also, when the information read from the encoded bit stream indicates that the FRUC mode is applied, the inter prediction unit 218 uses the pattern matching method (bilateral matching or template matching) read from the encoded stream. The motion information is derived by performing a motion search. Then, the inter prediction unit 218 performs motion compensation (prediction) using the derived motion information.

インター In addition, when the BIO mode is applied, the inter prediction unit 218 derives a motion vector based on a model assuming uniform linear motion. If the information read from the coded bit stream indicates that the affine motion compensation prediction mode is to be applied, the inter prediction unit 218 generates a motion vector in sub-block units based on motion vectors of a plurality of adjacent blocks. Is derived.

[MV derivation> Normal inter mode]
If the information read from the coded bit stream indicates that the normal inter mode is to be applied, the inter prediction unit 218 derives the MV based on the information read from the coded stream and uses the MV. To perform motion compensation (prediction).

FIG. 45 is a flowchart illustrating an example of inter prediction in the normal inter mode in the decoding device 200.

インター The inter prediction unit 218 of the decoding device 200 performs motion compensation on each block. The inter prediction unit 218 acquires a plurality of candidate MVs for the current block based on information such as the MVs of a plurality of decoded blocks around the current block temporally or spatially (step Ss_1). That is, the inter prediction unit 218 creates a candidate MV list.

Next, the inter prediction unit 218 assigns each of N (N is an integer of 2 or more) candidate MVs among a plurality of candidate MVs obtained in step Ss_1 to a motion vector predictor candidate (also referred to as a motion MV candidate). Are extracted according to a predetermined priority (step Ss_2). Note that the priority may be determined in advance for each of the N predicted MV candidates.

Next, the inter prediction unit 218 decodes the predicted motion vector selection information from the input stream (that is, the encoded bit stream), and uses the decoded prediction motion vector selection information to generate the N predicted MV candidates. Is selected as a predicted motion vector (also referred to as predicted MV) of the current block (step Ss_3).

Next, the inter prediction unit 218 decodes the difference MV from the input stream, and adds the difference value that is the decoded difference MV to the selected prediction motion vector, thereby obtaining the MV of the current block. It is derived (step Ss_4).

Finally, the inter prediction unit 218 generates a predicted image of the current block by performing motion compensation on the current block using the derived MV and the decoded reference picture (step Ss_5).

[Prediction control unit]
The prediction control unit 220 selects one of the intra prediction signal and the inter prediction signal, and outputs the selected signal to the addition unit 208 as a prediction signal. As a whole, the configuration, function, and processing of the prediction control unit 220, the intra prediction unit 216, and the inter prediction unit 218 on the decoding device side include the prediction control unit 128, the intra prediction unit 124, and the inter prediction unit 126 on the encoding device side. May correspond to the configuration, function, and processing.

[Decoding device implementation example]
FIG. 46 is a block diagram illustrating an implementation example of the decoding device 200. The decoding device 200 includes a processor b1 and a memory b2. For example, a plurality of components of the decoding device 200 illustrated in FIG. 41 are implemented by the processor b1 and the memory b2 illustrated in FIG.

The processor b1 is a circuit that performs information processing, and is a circuit that can access the memory b2. For example, the processor b1 is a dedicated or general-purpose electronic circuit that decodes an encoded moving image (that is, an encoded bit stream). The processor b1 may be a processor such as a CPU. Further, the processor b1 may be an aggregate of a plurality of electronic circuits. Further, for example, the processor b1 may play the role of a plurality of components among the plurality of components of the decoding device 200 illustrated in FIG. 41 and the like.

The memory b2 is a dedicated or general-purpose memory in which information for the processor b1 to decode the encoded bit stream is stored. The memory b2 may be an electronic circuit, and may be connected to the processor b1. Further, the memory b2 may be included in the processor b1. Further, the memory b2 may be an aggregate of a plurality of electronic circuits. The memory b2 may be a magnetic disk or an optical disk, or may be expressed as a storage or a recording medium. Further, the memory b2 may be a nonvolatile memory or a volatile memory.

{For example, the memory b2 may store a moving image or an encoded bit stream. Further, the memory b2 may store a program for the processor b1 to decode the encoded bit stream.

For example, the memory b2 may serve as a component for storing information among a plurality of components of the decoding device 200 illustrated in FIG. 41 and the like. Specifically, the memory b2 may serve as the block memory 210 and the frame memory 214 shown in FIG. More specifically, the memory b2 may store reconstructed blocks, reconstructed pictures, and the like.

In the decoding device 200, all of the plurality of components illustrated in FIG. 41 and the like may not be implemented, and all of the plurality of processes described above may not be performed. Some of the components illustrated in FIG. 41 and the like may be included in another device, or some of the above-described processes may be performed by another device.

[Definition of each term]
Each term may be defined as follows, for example.

A picture is an array of a plurality of luminance samples in a monochrome format, or an array of a plurality of luminance samples and a plurality of chrominance samples in a 4: 2: 0, 4: 2: 2 and 4: 4: 4 color format. This is the corresponding sequence. A picture may be a frame or a field.

The frame is a composition of a top field where a plurality of

sample rows

0, 2, 4,... Occur and a bottom field where a plurality of

sample rows

1, 3, 5,.

A slice is an integer number of coding trees contained in one independent slice segment and all subsequent dependent slice segments preceding the next independent slice segment (if any) in the same access unit (if any). Unit.

A tile is a rectangular area of a plurality of coding tree blocks in a specific tile column and a specific tile row in a picture. A tile may be a rectangular area of a frame, which is intended to be able to be decoded and coded independently, although a loop filter across the edges of the tile may still be applied.

The block is an M × N (N rows and M columns) array of a plurality of samples or an M × N array of a plurality of transform coefficients. A block may be a square or rectangular area of multiple pixels consisting of multiple matrices of one luminance and two color differences.

The CTU (coding tree unit) may be a coding tree block of a plurality of luminance samples of a picture having a three-sample arrangement, or may be two corresponding coding tree blocks of a plurality of chrominance samples. . Alternatively, the CTU is a multi-sample coded treeblock of any of a monochrome picture and a picture coded using three separate color planes and a syntax structure used to code the plurality of samples. It may be.

The super block may constitute one or two mode information blocks, or may be recursively divided into four 32 × 32 blocks and further divided into 64 × 64 pixel square blocks.

[First specific example of BIO]
Next, a first specific example of the decoding process based on the BIO will be described with reference to FIGS. 47, 48, and 49. For example, the decoding device 200 calculates a BIO parameter from two reference blocks on which motion compensation has been performed, and decodes the current block using the calculated BIO parameter.

FIG. 47 is a flowchart showing a first specific example of the decoding process based on BIO. FIG. 48 is a conceptual diagram illustrating an example of calculating a horizontal gradient value. FIG. 49 is a conceptual diagram illustrating a calculation example of a vertical gradient value.

First, the decoding apparatus 200 calculates a first sum for the current block using the horizontal gradient value of the first reference block and the horizontal gradient value of the second reference block (S1001). The current block may be a sub-block of the current coding unit (current CU) as shown in FIGS.

The first reference block is a block referred to in encoding the current block, and is a block defined in the reference picture L0 by the first motion vector of the current block or the current CU. The second reference block is a block referred to in encoding the current block, and is a block defined in the reference picture L1 by the second motion vector of the current block or the current CU.

Basically, the reference picture L0 and the reference picture L1 are two different reference pictures, and the first reference block and the second reference block are two different reference blocks. Further, here, a first reference block and a second reference block, each of which has been adjusted with a sub-pel precision by a motion compensation filter, are used, and have the same size as the current block.

If the current block is a sub-block of the current coding unit, each of the first reference block and the second reference block may be a sub-block of the reference block of the current coding unit.

That is, the plurality of pixel values in the reference picture L0 in FIGS. 48 and 49 may be a plurality of pixel values of a block determined as a reference block of the current encoding unit with respect to the reference picture L0. Similarly, the plurality of pixel values in the reference picture L1 in FIGS. 48 and 49 may be a plurality of pixel values of a block determined as a reference block of the current encoding unit with respect to the reference picture L1.

The decoding device 200 calculates the first sum using the above-described horizontal gradient value of the first reference block and the above-described horizontal gradient value of the second reference block. The decoding device 200 is not limited to the horizontal gradient value of the first reference block and the horizontal gradient value of the second reference block, but also the horizontal gradient value around the first reference block and the horizontal gradient value around the second reference block. The first sum may be calculated using the gradient value. The following equations (3.1) and (3.2) show examples of the first sum calculation processing method.

Here, × represents multiplication, and + represents addition. The sign represents a positive or negative sign, for example, the following.

As a _{result, sign (G x [i,} j]) × G x [i, j] _is, G x _[i, j] is directly _G x [i, j] is a positive value _{next, G} x [ If i, j] is a negative value, -G _x [i, j], which is equivalent to deriving the absolute value (abs) of G _x [i, j].

In the formula _{(3.1), I} ^{x 0} represents the horizontal gradient value in the first reference block of the reference picture _{L0, I} ^{x 1} represents the horizontal gradient value in the second reference block of the reference picture L1.

An example of a horizontal gradient filter for obtaining a horizontal gradient value is a 3-tap filter having a filter coefficient set of [-1,0,1]. The horizontal gradient value in the first reference block is calculated by applying a horizontal gradient filter to a plurality of reference pixels in the first reference block. The horizontal gradient value in the second reference block is calculated by applying a horizontal gradient filter to a plurality of reference pixels in the second reference block.

In the example shown in FIG. 48, the horizontal gradient value I _x ⁰ of the pixel located at [3, 2] of the first reference block is calculated as the matrix product [-1, ⁰ , 1] ^T [2, 3, 5]. And its value is 3. The horizontal gradient value I _x ¹ of the pixel located at [3, 2] in the second reference block is calculated as a matrix product [-1, 0, 1] ^T [5, 3, 2], and the value is −3. It is. [A, b, c] represents a matrix with three rows and one column.

In equation (3.2), sG _x represents the first sum and is calculated as the sum of absolute values of G _x over a window represented by Ω. The size of Ω may be the same as the current block. The size of Ω may be larger than the size of the current block. In the latter case, the value of G _x in adjacent positions of the current block is included in the calculation process of the first sum.

Further, the decoding device 200 calculates the second sum of the current block using the vertical gradient value of the first reference block and the vertical gradient value of the second reference block, similarly to the first sum which is the sum of the horizontal gradient values. It is calculated (S1002). The following equations (3.3) and (3.4) show examples of the second sum calculation processing method.

In Equation (3.3), I _y ⁰ represents a vertical gradient value in the first reference block of the reference picture L0, and I _y ¹ represents a vertical gradient value in the second reference block of the reference picture L1.

An example of a vertical gradient filter for obtaining a vertical gradient value is a three-tap filter having a filter coefficient set of [-1, 0, 1]. The vertical gradient value in the first reference block is calculated by applying a vertical gradient filter to a plurality of reference pixels in the first reference block. The vertical gradient value in the second reference block is calculated by applying a vertical gradient filter to a plurality of reference pixels in the second reference block.

In the example shown in FIG. 49, the vertical gradient value I _y ⁰ of the pixel located at [3, 2] of the first reference block is calculated as a matrix product [-1, ⁰ , 1] ^T [2, 3, 5]. And its value is 3. The vertical gradient value I _y ¹ of the pixel located at [3, 2] in the second reference block is calculated as the matrix product [-1, 0, 1] ^T [5, 3, 2], and the value is −3. It is.

In equation (3.4), sG _y represents the second sum and is calculated as the sum of absolute values of G _y over a window represented by Ω. If the size of Ω is larger than the size of the current block, the value of _{Gy at} a position adjacent to the current block is included in the second sum calculation process.

Next, the decoding device 200 determines whether or not the first sum is larger than the second sum (S1003). When it is determined that the first sum is larger than the second sum (Yes in S1003), the decoding device 200 determines a BIO parameter for the current block without using a vertical gradient value (S1004). Equations (3.5) to (3.9) show examples of arithmetic processing for determining the BIO parameters in this case. In these equations, the BIO parameter represented by u is calculated using the horizontal gradient value.

Here,-represents subtraction, and >> represents a shift operation. Bits, BIOSshift, and Clip respectively represent the following. In the following, “ceil” indicates a small number of round-ups, and “floor” indicates a small number of round-downs.

In equation (3.5), sG _x dI is calculated as the product sum of the difference between I ⁰ _{i, j} and I ¹ _{i, j} and −sign (G _x [i, j]) over the window Ω. . Here, I ⁰ _{i, j} represents a pixel value at a position [i, j] of the reference picture L0 in the first reference block, and I ¹ _{i, j} is a position of the reference picture L1 in the second reference block. Represents the pixel value at [i, j]. I ⁰ _{i, j} and I ¹ _{i, j} may be simply expressed as I ⁰ and I ¹ . The BIO parameter u is calculated using sG _x dI and sG _{x according} to equations (3.6) to (3.9).

場合 If it is determined that the first sum is not larger than the second sum (No in S1003), the decoding device 200 determines the BIO parameter u for the current block without using the horizontal gradient value (S1005). Equations (3.10) to (3.14) show examples of arithmetic processing for determining the BIO parameter u in this case. Equations (3.10) to (3.14) are basically the same as Equations (3.5) to (3.9), but in Equations (3.10) to (3.14), The BIO parameter u is calculated using the vertical gradient value.

In equation (3.10), sG _y dI is calculated as the product sum of the difference between I ⁰ _{i, j} and I ¹ _{i, j} and −sign (G _y [i, j]) over the window Ω. . The BIO parameter u is calculated using sG _y dI and sG _{y according} to equations (3.11) to (3.14).

{Then, the decoding device 200 decodes the current block using the BIO parameter u (S1006). Specifically, the decoding device 200 generates a prediction sample using the BIO parameter u, and decodes the current block using the prediction sample. Equations (3.15) and (3.16) show examples of arithmetic processing for generating a prediction sample.

場合 If it is determined that the first sum is larger than the second sum (Yes in S1003), equation (3.15) is used. When it is determined that the first sum is not larger than the second sum (No in S1003), Expression (3.16) is used.

The decoding apparatus 200 may repeat the above processing (S1001 to S1006) for all subblocks of the current CU.

The decoding device 200 can improve the accuracy of the prediction sample in the current block by using the BIO. Further, when calculating the BIO parameter, the decoding device 200 uses only one of the horizontal gradient value and the vertical gradient value, so that an increase in the amount of calculation can be suppressed.

Also, the above equation is an example, and the equation for calculating the BIO parameter is not limited to the above equation. For example, an equation equivalent to the above equation can be used. Also, the sign included in the above equation may be changed as appropriate. Specifically, the following equation (4.1) may be used as an equation corresponding to the above equations (3.1) and (3.2).

Also, for example, the following equation (4.2) may be used as an equation corresponding to the above equation (3.5).

Also, for example, the following equation (4.3) may be used as an equation corresponding to the above equation (3.15).

Expressions (3.6) to (3.9) substantially represent division, and therefore may be expressed as the following expression (4.4).

These equations (4.1) to (4.4) are equivalent to equations (3.1), (3.2), (3.5) to (3.9), and (3.15) described above. Substantially the same.

Similarly, specifically, the following equation (4.5) may be used as an equation corresponding to the above equations (3.3) and (3.4).

Also, for example, the following equation (4.6) may be used as an equation corresponding to the above equation (3.10).

Further, for example, the following equation (4.7) may be used as an equation corresponding to the above equation (3.16).

Expressions (3.11) to (3.14) substantially represent division, and therefore may be expressed as the following expression (4.8).

These equations (4.5) to (4.8) are equivalent to equations (3.3), (3.4), (3.10) to (3.14), and (3.16) described above. Substantially the same.

Also, in the above flow, the horizontal gradient value or the vertical gradient value is used based on the comparison between the first sum and the second sum, but the flow of the decoding process is not limited to the above flow. Whether to use the horizontal gradient value or the vertical gradient value may be determined in advance by another encoding parameter or the like. Then, the BIO parameter may be derived using the horizontal gradient value or the BIO parameter may be derived using the vertical gradient value without comparing the first sum and the second sum. Further, only one of the first sum and the second sum may be calculated.

Even if the first sum and the second sum are not compared, the decoding apparatus 200 can reduce the substantial multiplication that requires a large amount of calculation in the calculation performed for each pixel position by using the above-described equation. A plurality of parameters for generating an image can be derived with a small amount of calculation. Specifically, the decoding device 200 can calculate sG _x , sG _x dI, sG _y and sG _y dI with a small amount of calculation. Therefore, the decoding device 200 can reduce the amount of processing in decoding.

Although the decoding process is described above, the same process as described above can be applied to the encoding process. That is, the decoding in the above description may be replaced with encoding.

[Second Specific Example of BIO]
Next, a second specific example of the decoding process based on the BIO will be described. For example, similarly to the first specific example, the decoding device 200 calculates a BIO parameter from two reference blocks on which motion compensation has been performed, and decodes the current block using the calculated BIO parameter.

In this specific example, a prediction sample for decoding the current block is generated by the following equations (5.1) to (5.8).

_{S 1} of the formula (5.1) in the present embodiment corresponds to sG _x of formula (3.2) in the first embodiment. Also, _{S 2} of the formula (5.2) in the present embodiment corresponds to _sG x dI of formula (3.5) in the first embodiment. Further, _{S 3} of the formula (5.3) in the present embodiment corresponds to sG _y of formula (3.4) in the first embodiment. Further, _{S 4} of formula (5.4) in the present embodiment corresponds to _sG y dI of formula (3.10) in the first embodiment.

Then, _{v x} and _{v y} of the equation (5.5) - (5.7) in the present embodiment respectively correspond to BIO parameter, wherein in the first embodiment (3.9) and (3.14 ) Corresponds to u.

In the first specific example, in the calculation of sG _x , sG _x dI, sG _y and sG _y dI, the sign is changed, and no substantial multiplication is performed. On the other hand, in the present specific example, substantial multiplication is performed in the calculation of S ₁ , S ₂ , S _3, and S ₄ . This increases the amount of computation, but generates predicted samples with higher accuracy. Conversely, in the first specific example, the amount of calculation is reduced.

[Representative example of configuration and processing]
Representative examples of the configuration and processing of the encoding device 100 and the decoding device 200 described above are shown below. This representative example mainly corresponds to the first specific example described above.

FIG. 50 is a flowchart showing an operation performed by the encoding device 100. For example, the encoding device 100 includes a circuit and a memory connected to the circuit. The circuit and the memory included in the encoding device 100 may correspond to the processor a1 and the memory a2 illustrated in FIG. The circuit of the encoding device 100 performs the following in operation.

For example, the circuit of the encoding device 100 calculates, for each relative pixel position, the absolute value of the sum of the gradient value of the relative pixel position in the first range and the gradient value of the relative pixel position in the second range. A certain gradient sum absolute value is derived (S1101). Here, the first range includes a first reference block of the current block, and the second range includes a second reference block of the current block. Each relative pixel position is a pixel position that is commonly and relatively determined for both the first range and the second range, and is a pixel position in each of the first range and the second range. Here, being defined in common and relatively indicates that the pixel position in the first range and the pixel position in the second range used for obtaining one gradient sum absolute value are relatively the same. . For example, one gradient sum absolute value is derived using the gradient value for the pixel at the upper left position in the first range and the gradient value for the pixel at the upper left position in the second range.

回路 Then, the circuit of the encoding device 100 derives a total sum of a plurality of gradient sum absolute values derived for a plurality of relative pixel positions as a first parameter (S1102).

Further, the circuit of the encoding device 100 calculates, for each relative pixel position, a pixel difference which is a difference between the pixel value of the relative pixel position in the first range and the pixel value of the relative pixel position in the second range. A value is derived (S1103). For example, at this time, for each relative pixel position, the circuit of the encoding device 100 calculates the relative pixel position between the pixel value of the relative pixel position in the first range and the pixel value of the relative pixel position in the second range. By subtracting one from the other, a pixel difference value with a sign is derived.

{Circle around (1)} The circuit of the encoding device 100 inverts or maintains, for each relative pixel position, the sign of the pixel difference value derived for the relative pixel position using the sign of the gradient sum (S1104). Here, for each relative pixel position, the gradient sum is the sum of the gradient value of the relative pixel position in the first range and the gradient value of the relative pixel position in the second range.

{Circle around (1)} The circuit of the encoding device 100 derives, as the second parameter, the sum of the plurality of pixel difference values derived for each of the plurality of relative pixel positions and whose sign is inverted or maintained, respectively (S1105).

{Circle around (1)} The circuit of the encoding device 100 generates a predicted image used for encoding the current block using the first parameter and the second parameter (S1106).

Thereby, the encoding apparatus 100 can reduce substantial multiplication in which the amount of operation is large in the operation performed for each pixel position, and derive a plurality of parameters for generating a predicted image with a small amount of operation. Can be. Therefore, the encoding device 100 can reduce the amount of processing in encoding.

For example, for each relative pixel position, the gradient value of the relative pixel position in the first range may be the horizontal gradient value of the relative pixel position in the first range. Then, for each relative pixel position, the gradient value of the relative pixel position in the second range may be the horizontal gradient value of the relative pixel position in the second range. Thereby, the encoding device 100 can derive a plurality of parameters with a small amount of calculation using the horizontal gradient value.

{Also, for example, the circuit of the encoding device 100 may derive the first parameter by the above-described equation (4.1). Then, the circuit of the encoding device 100 may derive the second parameter by the above equation (4.2).

Here, Ω represents a set of a plurality of relative pixel positions, and [i, j] represents each relative pixel position. I _x ⁰ represents the horizontal gradient value of each relative pixel position in the first range, I _x ¹ represents the horizontal gradient value of each relative pixel position in the second range, and I ⁰ represents the first represents a pixel value of each relative pixel position in the range, I ¹ represents the pixel value of each relative pixel position in the second range. _{^{_{Further, abs (I x 1 + I}}} x 0) _represents the absolute value of ^{_{^{I x 1 + I x 0,}}} sign (I x 1 + I x 0) _represents the sign of ^_I ^x 1 ₊ ^I x ^0.

Thereby, the encoding device 100 can appropriately derive a plurality of parameters using the pixel value and the horizontal gradient value.

For example, for each relative pixel position, the gradient value of the relative pixel position in the first range may be a vertical gradient value of the relative pixel position in the first range. Then, for each relative pixel position, the gradient value of the relative pixel position in the second range may be a vertical gradient value of the relative pixel position in the second range. Thereby, the encoding apparatus 100 can derive a plurality of parameters with a small amount of calculation using the vertical gradient value.

{Also, for example, the circuit of the encoding device 100 may derive the first parameter by the above-described equation (4.5). Then, the circuit of the encoding device 100 may derive the second parameter by the above-described equation (4.6).

Here, Ω represents a set of a plurality of relative pixel positions, and [i, j] represents each relative pixel position. I _y ⁰ represents the vertical gradient value of each relative pixel position in the first range, I _y ¹ represents the vertical gradient value of each relative pixel position in the second range, and I ⁰ represents the first gradient value. represents a pixel value of each relative pixel position in the range, I ¹ represents the pixel value of each relative pixel position in the second range. _{^{_{Further, abs (I y 1 + I}}} y 0) _represents the absolute value of ^{_{^{I y 1 + I y 0,}}} sign (I y 1 + I y 0) _represents the sign of ^_I ^y 1 ₊ ^I y ^0.

Thereby, the encoding device 100 can appropriately derive a plurality of parameters using the pixel value and the vertical gradient value.

Also, for example, for each relative pixel position, the circuit of the encoding device 100 calculates the sum of the horizontal gradient value of the relative pixel position in the first range and the horizontal gradient value of the relative pixel position in the second range. May be derived as the absolute value of the horizontal gradient sum. Then, the circuit of the encoding device 100 may derive the total sum of the plurality of absolute values of the horizontal gradient sums derived for the plurality of relative pixel positions.

Further, for each relative pixel position, the circuit of the encoding device 100 calculates the absolute value of the sum of the vertical gradient value of the relative pixel position in the first range and the vertical gradient value of the relative pixel position in the second range. A vertical gradient sum absolute value that is a value may be derived. Then, the circuit of the encoding device 100 may derive the sum of the plurality of vertical gradient sum absolute values derived respectively for the plurality of relative pixel positions.

Here, when the sum of the plurality of absolute values of the horizontal gradient sums is greater than the sum of the absolute values of the plurality of vertical gradient sums, the circuit of the encoding device 100 may apply the horizontal gradient value to the gradient value. That is, in this case, the circuit of the encoding device 100 may apply the horizontal gradient value of the relative pixel position in the first range to the gradient value of each relative pixel position in the first range. In this case, the circuit of the encoding device 100 may apply the horizontal gradient value of the relative pixel position in the second range to the gradient value of each relative pixel position in the second range.

On the other hand, when the sum of the absolute values of the plurality of horizontal gradient sums is not greater than the sum of the absolute values of the plurality of vertical gradient sums, the circuit of the encoding device 100 may apply the vertical gradient value to the gradient value. That is, in this case, the circuit of the encoding device 100 may apply the vertical gradient value of the relative pixel position in the first range to the gradient value of each relative pixel position in the first range. Then, in this case, the circuit of the encoding device 100 may apply the vertical gradient value of the relative pixel position in the second range to the gradient value of each relative pixel position in the second range.

Thereby, the encoding apparatus 100 can apply the horizontal gradient value or the vertical gradient value to the derivation of the plurality of parameters based on the comparison between the sum related to the horizontal gradient value and the sum related to the vertical gradient value. it can. That is, the encoding device 100 can apply a gradient value having a large effect to the derivation of a plurality of parameters.

Further, for example, the circuit of the encoding device 100 derives a third parameter using the first parameter as a denominator and uses the second parameter as a numerator, and generates a predicted image using the third parameter. You may. Accordingly, the encoding device 100 can appropriately generate a predicted image using the parameters derived based on the relationship between the plurality of parameters.

In addition, for example, the circuit of the encoding device 100 uses a three-tap filter having a filter coefficient set of [−1, 0, 1] for each relative pixel position to generate a gradient value of the relative pixel position in the first range. , And the gradient value of the relative pixel position in the second range may be derived. Accordingly, the encoding device 100 can simply and appropriately derive the gradient value at each pixel position.

{Also, for example, the current block may be a sub-block defined by a 4 × 4 pixel size in the encoding unit. Then, the circuit of the encoding device 100 may derive the first parameter and the second parameter for each sub-block. Thereby, the encoding device 100 can appropriately generate a predicted image using the plurality of parameters determined by the 4 × 4 pixel size.

Also, for example, the circuit of the encoding device 100 derives parameters of the bidirectional optical flow using the first parameter and the second parameter, and performs prediction using the parameters of the bidirectional optical flow and the bidirectional optical flow. An image may be generated. Thereby, the encoding device 100 can appropriately generate a predicted image. The parameter of the bidirectional optical flow may be the third parameter described above.

Note that the inter prediction unit 126 of the encoding device 100 may perform the above-described operation as a circuit of the encoding device 100.

FIG. 51 is a flowchart showing an operation performed by the decoding device 200. For example, the decoding device 200 includes a circuit and a memory connected to the circuit. The circuit and the memory included in the decoding device 200 may correspond to the processor b1 and the memory b2 illustrated in FIG. The circuit of the decoding device 200 performs the following in operation.

For example, the circuit of the decoding device 200 calculates, for each relative pixel position, the absolute value of the sum of the gradient value of the relative pixel position in the first range and the gradient value of the relative pixel position in the second range. The gradient sum absolute value is derived (S1201). Here, the first range includes a first reference block of the current block, and the second range includes a second reference block of the current block. Each relative pixel position is a pixel position that is commonly and relatively determined for both the first range and the second range, and is a pixel position in each of the first range and the second range.

回路 Then, the circuit of the decoding device 200 derives, as the first parameter, the sum total of the plurality of gradient sum absolute values derived for the plurality of relative pixel positions (S1202).

Further, the circuit of the decoding device 200 calculates, for each relative pixel position, a pixel difference value that is a difference between the pixel value of the relative pixel position in the first range and the pixel value of the relative pixel position in the second range. Is derived (S1203). For example, at this time, for each relative pixel position, the circuit of the decoding device 200 determines one of the pixel value of the relative pixel position in the first range and the pixel value of the relative pixel position in the second range. By subtracting the other from, a pixel difference value with a sign is derived.

{Circle around (2)} The circuit of the decoding device 200 inverts or maintains the sign of the pixel difference value derived for the relative pixel position by the sign of the gradient sum for each relative pixel position (S1204). Here, for each relative pixel position, the gradient sum is the sum of the gradient value of the relative pixel position in the first range and the gradient value of the relative pixel position in the second range.

{Circle around (2)} The circuit of the decoding device 200 derives, as a second parameter, the sum of the plurality of pixel difference values derived for each of the plurality of relative pixel positions and whose sign is inverted or maintained (S1205).

Then, the circuit of the decoding device 200 generates a predicted image used for decoding the current block using the first parameter and the second parameter (S1206).

Thereby, the decoding apparatus 200 can reduce substantial multiplication in which the calculation amount is large in the calculation performed for each pixel position, and can derive a plurality of parameters for generating a predicted image with a small calculation amount. it can. Therefore, the decoding device 200 can reduce the amount of processing in decoding.

For example, for each relative pixel position, the gradient value of the relative pixel position in the first range may be the horizontal gradient value of the relative pixel position in the first range. Then, for each relative pixel position, the gradient value of the relative pixel position in the second range may be the horizontal gradient value of the relative pixel position in the second range. Thereby, the decoding device 200 can derive a plurality of parameters with a small amount of calculation using the horizontal gradient value.

{Also, for example, the circuit of the decoding device 200 may derive the first parameter by the above-described equation (4.1). Then, the circuit of the decoding device 200 may derive the second parameter by the above equation (4.2).

Here, Ω represents a set of a plurality of relative pixel positions, and [i, j] represents each relative pixel position. I _x ⁰ represents the horizontal gradient value of each relative pixel position in the first range, I _x ¹ represents the horizontal gradient value of each relative pixel position in the second range, and I ⁰ represents the first gradient value. represents a pixel value of each relative pixel position in the range, I ¹ represents the pixel value of each relative pixel position in the second range. _{^{_{Further, abs (I x 1 + I}}} x 0) _represents the absolute value of ^{_{^{I x 1 + I x 0,}}} sign (I x 1 + I x 0) _represents the sign of ^_I ^x 1 ₊ ^I x ^0.

Thereby, the decoding device 200 can appropriately derive a plurality of parameters using the pixel value and the horizontal gradient value.

For example, for each relative pixel position, the gradient value of the relative pixel position in the first range may be a vertical gradient value of the relative pixel position in the first range. Then, for each relative pixel position, the gradient value of the relative pixel position in the second range may be a vertical gradient value of the relative pixel position in the second range. Accordingly, the decoding device 200 can derive a plurality of parameters with a small amount of calculation using the vertical gradient value.

{Also, for example, the circuit of the decoding device 200 may derive the first parameter by the above-described equation (4.5). Then, the circuit of the decoding device 200 may derive the second parameter by the above equation (4.6).

Here, Ω represents a set of a plurality of relative pixel positions, and [i, j] represents each relative pixel position. I _y ¹ represents the vertical gradient value of each relative pixel position in the second range, I ⁰ represents the pixel value of each relative pixel position in the first range, and I ¹ represents each relative pixel position in the second range. Represents the pixel value at the target pixel position. _{^{_{Further, abs (I y 1 + I}}} y 0) _represents the absolute value of ^{_{^{I y 1 + I y 0,}}} sign (I y 1 + I y 0) _represents the sign of ^_I ^y 1 ₊ ^I y ^0. And I _y ⁰ represents the vertical gradient value of each relative pixel position in the first range.

Thereby, the decoding device 200 can appropriately derive a plurality of parameters using the pixel value and the vertical gradient value.

Further, for example, for each relative pixel position, the circuit of the decoding device 200 calculates the sum of the horizontal gradient value of the relative pixel position in the first range and the horizontal gradient value of the relative pixel position in the second range. A horizontal gradient sum absolute value that is an absolute value may be derived. Then, the circuit of the decoding device 200 may derive the sum of the plurality of horizontal gradient sum absolute values derived for the plurality of relative pixel positions.

Further, the circuit of the decoding device 200 calculates, for each relative pixel position, the absolute value of the sum of the vertical gradient value of the relative pixel position in the first range and the vertical gradient value of the relative pixel position in the second range. May be derived. Then, the circuit of the decoding device 200 may derive the total sum of the plurality of vertical gradient sum absolute values derived for each of the plurality of relative pixel positions.

Here, when the sum of the plurality of absolute values of the horizontal gradient sums is larger than the sum of the absolute values of the plurality of vertical gradient sums, the circuit of the decoding device 200 may apply the horizontal gradient value to the gradient value. That is, in this case, the circuit of the decoding device 200 may apply the horizontal gradient value of the relative pixel position in the first range to the gradient value of each relative pixel position in the first range. In this case, the circuit of the decoding device 200 may apply the horizontal gradient value of the relative pixel position in the second range to the gradient value of each relative pixel position in the second range.

On the other hand, when the sum of the absolute values of the plurality of horizontal gradient sums is not greater than the sum of the absolute values of the plurality of vertical gradient sums, the circuit of the decoding device 200 may apply the vertical gradient value to the gradient value. That is, in this case, the circuit of the decoding device 200 may apply the vertical gradient value of the relative pixel position in the first range to the gradient value of each relative pixel position in the first range. Then, in this case, the circuit of the decoding device 200 may apply the vertical gradient value of the relative pixel position in the second range to the gradient value of each relative pixel position in the second range.

Thereby, the decoding device 200 can apply the horizontal gradient value or the vertical gradient value to the derivation of the plurality of parameters based on the comparison between the sum related to the horizontal gradient value and the sum related to the vertical gradient value. . That is, the decoding device 200 can apply a gradient value having a large effect to the derivation of a plurality of parameters.

Also, for example, the circuit of the decoding device 200 derives a third parameter using the first parameter as a denominator and uses the second parameter as a numerator, and generates a prediction image using the third parameter. Is also good. Thereby, the decoding device 200 can appropriately generate a predicted image using the parameters derived based on the relationship between the plurality of parameters.

In addition, for example, the circuit of the decoding device 200 uses a three-tap filter having a filter coefficient set of [−1, 0, 1] for each relative pixel position to calculate the gradient value of the relative pixel position in the first range. Further, a gradient value of the relative pixel position in the second range may be derived. Thus, the decoding device 200 can simply and appropriately derive the gradient value at each pixel position.

{Also, for example, the current block may be a sub-block defined by a 4 × 4 pixel size in the encoding unit. Then, the circuit of the decoding device 200 may derive the first parameter and the second parameter for each sub-block. Thus, the decoding device 200 can appropriately generate a predicted image using a plurality of parameters determined by the 4 × 4 pixel size.

Also, for example, the circuit of the decoding device 200 derives the parameters of the bidirectional optical flow using the first parameter and the second parameter, and uses the parameters of the bidirectional optical flow and the bidirectional optical flow to generate a predicted image. May be generated. Thereby, the decoding device 200 can appropriately generate a predicted image. The parameter of the bidirectional optical flow may be the third parameter described above.

Note that the inter prediction unit 218 of the decoding device 200 may perform the above-described operation as a circuit of the decoding device 200.

[Other examples]
The encoding device 100 and the decoding device 200 in each example described above may be used as an image encoding device and an image decoding device, respectively, or may be used as a moving image encoding device and a moving image decoding device. .

Alternatively, each of the encoding device 100 and the decoding device 200 may be used as a prediction device. That is, the encoding device 100 and the decoding device 200 may correspond to only the inter prediction unit 126 and the inter prediction unit 218, respectively. And other components may be included in other devices.

In addition, at least a part of each example described above may be used as an encoding method, may be used as a decoding method, may be used as a prediction method, and may be used as another method. You may.

Each component may be configured by dedicated hardware, or may be realized by executing a software program suitable for each component. Each component may be realized by a program execution unit such as a CPU or a processor reading and executing a software program recorded on a recording medium such as a hard disk or a semiconductor memory.

Specifically, each of the encoding device 100 and the decoding device 200 includes a processing circuit (Processing @ Circuitry) and a storage device (Storage) electrically connected to the processing circuit and accessible from the processing circuit. You may have. For example, the processing circuit corresponds to the processor a1 or b1, and the storage device corresponds to the memory a2 or b2.

(4) The processing circuit includes at least one of dedicated hardware and a program execution unit, and executes processing using a storage device. When the processing circuit includes a program execution unit, the storage device stores a software program executed by the program execution unit.

Here, the software that implements the above-described encoding device 100 or decoding device 200 is the following program.

For example, the program may cause the computer to determine a plurality of common and relatively defined regions for both a first range including a first reference block of the current block and a second range including a second reference block of the current block. For each of a plurality of relative pixel positions that are a plurality of pixel positions in each of the first range and the second range, a gradient value of the relative pixel position in the first range, Deriving a gradient sum absolute value that is an absolute value of the sum of the relative pixel position and the gradient value in the second range, and calculating the sum of the plurality of gradient sum absolute values respectively derived for the plurality of relative pixel positions. One parameter, and for each of the plurality of relative pixel positions, the pixel value of the relative pixel position in the first range and the second range , A pixel difference value that is a difference between the pixel value of the relative pixel position and the gradient value of the relative pixel position in the first range is calculated for each of the plurality of relative pixel positions. By the sign of the gradient sum that is the sum of the gradient value of the relative pixel position in the range, the sign of the pixel difference value derived for the relative pixel position is inverted or maintained, and the plurality of relative pixels The sum of a plurality of pixel difference values, each of which is derived for a position and the sign of which is inverted or maintained, is derived as a second parameter, and the code of the current block is obtained using the first parameter and the second parameter. Encoding method for generating a predicted image used for image conversion.

In addition, for example, the program defines in the computer both a first range including the first reference block of the current block and a second range including the second reference block of the current block, which are common and relatively defined. A plurality of relative pixel positions, which are a plurality of pixel positions in the first range and the second range, respectively, and a gradient value of the relative pixel position in the first range. Deriving a gradient sum absolute value, which is an absolute value of a sum of the gradient value of the relative pixel position and the gradient value in the second range, and summing a plurality of gradient sum absolute values respectively derived for the plurality of relative pixel positions. As a first parameter, and for each of the plurality of relative pixel positions, a pixel value of the relative pixel position in the first range, Deriving a pixel difference value that is a difference between the pixel value of the relative pixel position in the two ranges and the gradient value of the relative pixel position in the first range for each of the plurality of relative pixel positions; The sign of the pixel difference value derived for the relative pixel position is inverted or maintained by the sign of the gradient sum that is the sum of the gradient value of the relative pixel position in the second range and the plurality of relative values. The second block derives, as a second parameter, a sum of a plurality of pixel difference values, each of which is derived for a target pixel position and whose sign is inverted or maintained, and uses the first parameter and the second parameter to calculate the current block. May be executed by executing a decoding method for generating a predicted image used for decoding of the image.

Each component may be a circuit as described above. These circuits may constitute one circuit as a whole, or may be separate circuits. Further, each component may be realized by a general-purpose processor, or may be realized by a dedicated processor.

処理 Alternatively, a process performed by a specific component may be performed by another component. The order in which the processes are performed may be changed, or a plurality of processes may be performed in parallel. Further, the encoding / decoding device may include the encoding device 100 and the decoding device 200.

The ordinal numbers such as the first and second ordinal numbers used in the description may be appropriately changed. Also, ordinal numbers may be newly given to components or the like, or may be removed.

As described above, the aspects of the encoding device 100 and the decoding device 200 have been described based on a plurality of examples, but the aspects of the encoding device 100 and the decoding device 200 are not limited to these examples. Unless departing from the gist of the present disclosure, various modifications conceivable by those skilled in the art may be applied to each example, and a configuration constructed by combining components in different examples may be in the range of the aspects of the encoding apparatus 100 and the decoding apparatus 200. May be included within.

１One or more aspects disclosed herein may be implemented in combination with at least a part of other aspects of the present disclosure. In addition, a part of the processing, a part of the configuration of the device, a part of the syntax, and the like described in the flowchart of one or more aspects disclosed herein may be implemented in combination with another aspect.

[Implementation and application]
In each of the above embodiments, each of the functional or functional blocks can be generally realized by an MPU (micro processing unit), a memory, and the like. Further, the processing by each of the functional blocks may be realized as a program execution unit such as a processor that reads and executes software (program) recorded on a recording medium such as a ROM. The software may be distributed. The software may be recorded on various recording media such as a semiconductor memory. Each functional block can be realized by hardware (dedicated circuit). Various combinations of hardware and software may be employed.

The processing described in each embodiment may be realized by centralized processing using a single device (system), or may be realized by distributed processing using a plurality of devices. In addition, the number of processors that execute the program may be one or more. That is, centralized processing or distributed processing may be performed.

態様 Aspects of the present disclosure are not limited to the embodiments described above, and various modifications are possible, and these are also included in the scope of the aspects of the present disclosure.

Further, here, application examples of the moving picture coding method (image coding method) or the moving picture decoding method (image decoding method) described in each of the above embodiments, and various systems for implementing the applied examples are described. explain. Such a system may be characterized by having an image encoding device using an image encoding method, an image decoding device using an image decoding method, or an image encoding / decoding device including both. Other configurations of such a system can be appropriately changed as necessary.

[Example of use]
FIG. 52 is a diagram illustrating an overall configuration of an appropriate content supply system ex100 that realizes a content distribution service. A communication service providing area is divided into desired sizes, and base stations ex106, ex107, ex108, ex109, and ex110, which are fixed wireless stations in the illustrated example, are installed in each cell.

In this content supply system ex100, each device such as a computer ex111, a game machine ex112, a camera ex113, a home appliance ex114, and a smartphone ex115 is connected to the Internet ex101 via the Internet service provider ex102 or the communication network ex104 and the base stations ex106 to ex110. Is connected. The content supply system ex100 may be connected by combining any of the above devices. In various implementations, each device may be directly or indirectly interconnected via a telephone network or short-range wireless communication without using the base stations ex106 to ex110. Further, the streaming server ex103 may be connected to each device such as the computer ex111, the game machine ex112, the camera ex113, the home appliance ex114, and the smartphone ex115 via the Internet ex101 and the like. Further, the streaming server ex103 may be connected to a terminal or the like in a hot spot in the airplane ex117 via the satellite ex116.

Instead of the base stations ex106 to ex110, a wireless access point or a hot spot may be used. Further, the streaming server ex103 may be directly connected to the communication network ex104 without going through the Internet ex101 or the Internet service provider ex102, or may be directly connected to the airplane ex117 without going through the satellite ex116.

The camera ex113 is a device such as a digital camera capable of capturing a still image and a moving image. The smartphone ex115 is a smartphone, a mobile phone, a PHS (Personal Handy-phone System), or the like that supports a mobile communication system called 2G, 3G, 3.9G, 4G, and 5G in the future.

The home appliance ex114 is a device included in a refrigerator or a home fuel cell cogeneration system.

In the content supply system ex100, a terminal having a photographing function is connected to the streaming server ex103 via the base station ex106 or the like, thereby enabling live distribution or the like. In the live distribution, the terminal (computer ex111, game machine ex112, camera ex113, home appliance ex114, smartphone ex115, terminal in the airplane ex117, etc.) performs the above-described processing on the still or moving image content shot by the user using the terminal. The encoding process described in each embodiment may be performed, and the video data obtained by the encoding may be multiplexed with the audio data obtained by encoding the sound corresponding to the video, and the obtained data may be streamed. It may be transmitted to the server ex103. That is, each terminal functions as an image encoding device according to an aspect of the present disclosure.

On the other hand, the streaming server ex103 stream-distributes the transmitted content data to the requested client. The client is a computer ex111, a game machine ex112, a camera ex113, a home appliance ex114, a smartphone ex115, a terminal in an airplane ex117, or the like, which can decode the encoded data. Each device that has received the distributed data may decode the received data and reproduce it. That is, each device may function as the image decoding device according to an aspect of the present disclosure.

[Distributed processing]
The streaming server ex103 may be a plurality of servers or a plurality of computers, and may process, record, or distribute data in a distributed manner. For example, the streaming server ex103 may be realized by a CDN (Contents Delivery Network), and the content distribution may be realized by a large number of edge servers distributed around the world and a network connecting the edge servers. In a CDN, physically close edge servers can be dynamically assigned depending on the client. Then, the delay can be reduced by caching and distributing the content to the edge server. In addition, when some types of errors occur or when the communication state changes due to an increase in traffic, the processing is distributed among a plurality of edge servers, the distribution entity is switched to another edge server, or a failure occurs. Since the distribution can be continued by bypassing the network, high-speed and stable distribution can be realized.

In addition to the distributed processing of the distribution itself, the encoding processing of the captured data may be performed by each terminal, may be performed on the server side, or may be performed by sharing with each other. As an example, generally, in an encoding process, a processing loop is performed twice. In the first loop, the complexity or code amount of an image in units of frames or scenes is detected. In the second loop, processing for maintaining the image quality and improving the coding efficiency is performed. For example, the terminal performs the first encoding process, and the server that has received the content performs the second encoding process, thereby improving the quality and efficiency of the content while reducing the processing load on each terminal. it can. In this case, if there is a request to receive and decode in near real time, the first encoded data performed by the terminal can be received and played back by another terminal, so more flexible real time distribution is possible Become.

As another example, the camera ex113 or the like extracts a feature amount (feature or characteristic amount) from an image, compresses data related to the feature amount as metadata, and transmits the metadata to the server. The server performs compression according to the meaning of the image (or the importance of the content), such as switching the quantization accuracy by determining the importance of the object from the feature amount. The feature amount data is particularly effective for improving the accuracy and efficiency of motion vector prediction at the time of recompression at the server. Alternatively, the terminal may perform simple coding such as VLC (variable length coding), and the server may perform coding with a large processing load such as CABAC (context-adaptive binary arithmetic coding).

As yet another example, in a stadium, a shopping mall, a factory, or the like, a plurality of terminals may have a plurality of video data obtained by shooting substantially the same scene. In this case, for example, a GOP (Group of Picture) unit, a picture unit, or a tile obtained by dividing a picture is used by using a plurality of terminals that have taken a picture and, if necessary, other terminals and servers that have not taken a picture. Distributed processing is performed by assigning encoding processing in units or the like. Thereby, delay can be reduced and more real-time properties can be realized.

Since a plurality of video data are almost the same scene, the server may manage and / or give an instruction so that video data shot by each terminal can be referred to each other. Further, the encoded data from each terminal may be received by the server, and the reference relationship may be changed between a plurality of data, or the picture itself may be corrected or replaced to be re-encoded. As a result, it is possible to generate a stream in which the quality and efficiency of each data is improved.

{Circle around (2)} The server may perform the transcoding for changing the encoding method of the video data, and then distribute the video data. For example, the server may convert an MPEG-based coding method to a VP-based (for example, VP9) or H.264-based. H.264 to H.264. 265.

As described above, the encoding process can be performed by the terminal or one or more servers. Therefore, in the following, description such as “server” or “terminal” will be used as the subject of processing, but part or all of the processing performed by the server may be performed by the terminal, or the processing performed by the terminal may be performed. Some or all may be performed at the server. The same applies to the decoding process.

[3D, multi-angle]
Different scenes captured by a plurality of terminals such as the camera ex113 and / or the smartphone ex115, which are substantially synchronized with each other, or images or videos of the same scene captured from different angles have been increasingly integrated and used. The video shot by each terminal can be integrated based on the relative positional relationship between the separately obtained terminals, the region where the feature points included in the video match, and the like.

The server not only encodes a two-dimensional moving image, but also automatically encodes a still image based on scene analysis of the moving image or at a time designated by the user and transmits the encoded still image to the receiving terminal. Is also good. If the server can further acquire the relative positional relationship between the photographing terminals, the server can change the three-dimensional shape of the scene based on not only a two-dimensional moving image but also a video of the same scene photographed from different angles. Can be generated. The server may separately encode the three-dimensional data generated by the point cloud or the like, or generate a plurality of images to be transmitted to the receiving terminal based on the result of recognizing or tracking a person or an object using the three-dimensional data. May be selected or reconstructed from the video taken by the terminal.

In this way, the user can arbitrarily select each video corresponding to each photographing terminal to enjoy the scene, and can select the video of the selected viewpoint from the three-dimensional data reconstructed using a plurality of images or videos. You can also enjoy clipped content. Further, along with the video, the sound is collected from a plurality of different angles, and the server may multiplex the sound from a specific angle or space with the corresponding video, and transmit the multiplexed video and sound. Good.

In recent years, contents in which the real world is associated with the virtual world, such as Virtual Reality (VR) and Augmented Reality (AR), have become widespread. In the case of a VR image, the server may create right-eye and left-eye viewpoint images, and perform encoding that allows reference between viewpoint images by Multi-View @ Coding (MVC) or the like. It may be encoded as a separate stream without reference. At the time of decoding another stream, it is preferable that the streams are reproduced in synchronization with each other so that a virtual three-dimensional space is reproduced according to the viewpoint of the user.

In the case of the AR image, the server may superimpose the virtual object information in the virtual space on the camera information in the real space based on the three-dimensional position or the movement of the viewpoint of the user. The decoding device may obtain or hold the virtual object information and the three-dimensional data, generate a two-dimensional image according to the movement of the user's viewpoint, and create the superimposition data by connecting the two-dimensional images smoothly. Alternatively, the decoding device may transmit the movement of the user's viewpoint to the server in addition to the request for the virtual object information. The server may create superimposed data in accordance with the movement of the viewpoint received from the three-dimensional data stored in the server, encode the superimposed data, and distribute the encoded data to the decoding device. Note that the superimposed data typically has an α value indicating transparency other than RGB, and the server sets the α value of a portion other than the object created from the three-dimensional data to 0 or the like, and May be encoded in a state where is transmitted. Alternatively, the server may generate data in which a predetermined RGB value such as a chroma key is set as a background, and a portion other than the object is set as a background color. The predetermined RGB values may be determined in advance.

Similarly, the decoding process of the distributed data may be performed by a client (for example, a terminal), may be performed by a server, or may be performed by sharing with each other. As an example, a certain terminal may once send a reception request to the server, receive the content corresponding to the request by another terminal, perform a decoding process, and transmit a decoded signal to a device having a display. Data with good image quality can be reproduced by selecting an appropriate content by dispersing the processing regardless of the performance of the communicable terminal itself. As another example, while receiving large-size image data by a TV or the like, a partial area such as a tile obtained by dividing a picture may be decoded and displayed on a personal terminal of a viewer. As a result, while sharing the entire image, it is possible to check at hand the field in which the user is responsible or the area that the user wants to check in more detail.

コンテンツ In situations where multiple indoor, outdoor, short-range, medium-range or long-range wireless communications can be used, it may be possible to seamlessly receive content using a distribution system standard such as MPEG-DASH. The user may switch in real time while freely selecting a user's terminal, a decoding device such as a display placed indoors or outdoors, or a display device. In addition, decoding can be performed while switching between a terminal to be decoded and a terminal to be displayed using the own position information or the like. Thereby, while the user is moving to the destination, it is possible to map and display information on a wall of a neighboring building or a part of the ground where the displayable device is embedded. Access to encoded data on a network, such as when encoded data is cached on a server that can be accessed from a receiving terminal in a short time, or copied to an edge server in a content delivery service. It is also possible to switch the bit rate of the received data based on ease.

[Scalable encoding]
Switching of contents will be described using a scalable stream that is compression-encoded by applying the moving image encoding method described in each of the above embodiments and illustrated in FIG. The server may have a plurality of streams having the same content and different qualities as individual streams, but the temporal / spatial scalable realization is realized by performing encoding by dividing into layers as shown in the figure. A configuration in which the content is switched by utilizing the characteristics of the stream may be employed. In other words, the decoding side determines which layer to decode according to an internal factor such as performance and an external factor such as a communication band state, so that the decoding side can separate low-resolution content and high-resolution content. You can switch freely to decode. For example, when the user wants to watch the continuation of the video that was being viewed on the smartphone ex115 while moving, for example, after returning home, using a device such as an Internet TV, the device only needs to decode the same stream to a different layer. The burden on the side can be reduced.

Furthermore, as described above, the picture is encoded for each layer, and in addition to the configuration that realizes scalability in the enhancement layer above the base layer, the enhancement layer includes meta information based on image statistical information and the like. Is also good. The decoding side may generate high-quality content by super-resolution of the base layer picture based on the meta information. Super-resolution may improve the signal-to-noise ratio while maintaining and / or enlarging the resolution. Meta information is information for specifying a linear or non-linear filter coefficient used for super-resolution processing, or information for specifying a parameter value in filter processing, machine learning, or least-squares operation used for super-resolution processing, and the like. including.

Alternatively, a configuration in which a picture is divided into tiles or the like according to the meaning of an object or the like in an image may be provided. The decoding side decodes only a partial area by selecting a tile to be decoded. Furthermore, by storing the attribute of the object (person, car, ball, etc.) and the position in the video (coordinate position in the same image, etc.) as meta information, the decoding side can determine the position of the desired object based on the meta information. , And the tile that contains the object can be determined. For example, as shown in FIG. 54, the meta information may be stored using a data storage structure different from the pixel data, such as a SEI (supplemental enhancement information) message in HEVC. This meta information indicates, for example, the position, size, color, etc. of the main object.

メタ Meta information may be stored in a unit composed of a plurality of pictures, such as a stream, a sequence, or a random access unit. The decoding side can acquire the time at which the specific person appears in the video, and can determine the picture in which the object exists by combining the information in the picture unit with the time information, and can determine the position of the object in the picture.

[Web page optimization]
FIG. 55 is a diagram illustrating an example of a display screen of a web page on the computer ex111 or the like. FIG. 56 is a diagram illustrating a display screen example of a web page on the smartphone ex115 or the like. As shown in FIGS. 55 and 56, a web page may include a plurality of link images, which are links to image contents, and the appearance may differ depending on a viewing device. When a plurality of link images can be seen on the screen, the display device (until the link image approaches the center of the screen or the entire link image enters the screen until the user explicitly selects the link image) The decoding device) may display a still image or an I picture included in each content as a link image, may display a video such as a gif animation with a plurality of still images or I pictures, or may include a base layer. , And may decode and display the video.

(4) When the link image is selected by the user, the display device performs decoding while giving priority to the base layer, for example. If there is information indicating that the content is scalable in the HTML constituting the web page, the display device may decode the content up to the enhancement layer. Furthermore, in order to ensure real-time performance, before the selection or when the communication band is extremely severe, the display device decodes only forward-referenced pictures (I-pictures, P-pictures, and B-pictures with only forward-reference). And display, the delay between the decoding time of the first picture and the display time (the delay from the start of the decoding of the content to the start of the display) can be reduced. Still further, the display device may intentionally ignore the reference relation of pictures, perform coarse decoding with all B pictures and P pictures being forward-referenced, and perform normal decoding as time passes and more pictures are received. .

[Automatic driving]
Further, when transmitting and receiving still image or video data such as two-dimensional or three-dimensional map information for automatic driving or driving support of a car, the receiving terminal is required to transmit meta data in addition to image data belonging to one or more layers. Weather or construction information may also be received as information, and these may be associated and decoded. Note that the meta information may belong to a layer or may be simply multiplexed with image data.

In this case, since a car, a drone or an airplane including the receiving terminal moves, the receiving terminal transmits the position information of the receiving terminal, and performs seamless reception and decoding while switching between the base stations ex106 to ex110. realizable. Further, the receiving terminal dynamically switches how much the meta information is received or how much the map information is updated according to the selection of the user, the status of the user, and / or the state of the communication band. Becomes possible.

In the content supply system ex100, the client can receive, decode, and reproduce the encoded information transmitted by the user in real time.

[Distribution of personal content]
Further, in the content supply system ex100, not only high-quality and long-time content by a video distributor but also low-quality and short-time content unicast or multicast distribution by an individual is possible. Such personal content is expected to increase in the future. In order to make the personal content better, the server may perform the encoding process after performing the editing process. This can be realized using, for example, the following configuration.

(4) The server performs a recognition process such as a shooting error, a scene search, a meaning analysis, and an object detection from the original image data or the encoded data after shooting in real time or after storing and shooting. Then, the server manually or automatically corrects out-of-focus or camera shake based on the recognition result, or deletes a less important scene such as a scene whose brightness is lower or out of focus compared to other pictures. Perform editing such as deleting, emphasizing the edges of the object, and changing the color. The server encodes the edited data based on the editing result. It is also known that if the shooting time is too long, the audience rating will decrease, and the server will not only move scenes with low importance as described above so that the content will be within a specific time range according to the shooting time, but also May be automatically clipped based on the image processing result. Alternatively, the server may generate and encode the digest based on the result of the semantic analysis of the scene.

In some cases, personal content includes infringement of copyright, moral rights, or portrait right, etc., which is inconvenient for individuals, such as the range of sharing exceeding the intended range. There is also. Therefore, for example, the server may dare to change the image of a person's face in the periphery of the screen or the inside of a house into an image out of focus. Further, the server recognizes whether or not a face of a person different from the person registered in advance is shown in the image to be encoded, and if so, performs processing such as mosaicing the face part. You may. Alternatively, as the pre-processing or post-processing of the encoding, the user may designate a person or a background area where the user wants to process the image from the viewpoint of copyright or the like. The server may perform processing such as replacing the designated area with another image or defocusing. If it is a person, it is possible to track the person in the moving image and replace the image of the face of the person.

視聴 Since the viewing of personal contents with a small data amount requires a real-time property, the decoding apparatus may first receive the base layer with the highest priority to perform decoding and playback, depending on the bandwidth. The decoding device may receive the enhancement layer during this time, and reproduce the high-quality video including the enhancement layer when the reproduction is performed twice or more, such as when the reproduction is looped. If the stream is scalable encoded in this way, it is a rough moving image when not selected or when it is started to be viewed, but it is possible to provide an experience in which the stream becomes smarter and the image improves gradually. In addition to the scalable encoding, a similar experience can be provided even if the coarse stream reproduced at the first time and the second stream encoded with reference to the first moving image are configured as one stream. .

[Other application examples]
In addition, these encoding or decoding processes are generally performed in the LSI ex500 included in each terminal. The LSI (large scale integration circuit) ex500 (see FIG. 52) may be a single chip or a configuration including a plurality of chips. It should be noted that the video encoding or decoding software is incorporated into any recording medium (CD-ROM, flexible disk, hard disk, or the like) readable by the computer ex111 or the like, and encoding or decoding processing is performed using the software. Is also good. Further, when the smartphone ex115 has a camera, the moving image data acquired by the camera may be transmitted. The moving image data at this time may be data encoded by the LSI ex500 of the smartphone ex115.

Note that the LSI ex500 may be configured to download and activate application software. In this case, the terminal first determines whether the terminal supports the content encoding method or has the ability to execute the specific service. If the terminal does not support the content encoding method or does not have a specific service execution capability, the terminal may download a codec or application software, and then acquire and reproduce the content.

In addition to the content supply system ex100 via the Internet ex101, at least the moving picture coding apparatus (picture coding apparatus) or the moving picture decoding apparatus (picture decoding apparatus) of each of the above-described embodiments is applicable to a digital broadcasting system. Can be incorporated. Since the multiplexed data in which video and sound are multiplexed on a radio wave for broadcasting using a satellite or the like is transmitted and received, there is a difference that the configuration of the content supply system ex100 is suitable for multicasting, in contrast to the configuration that facilitates unicasting. However, similar applications are possible for the encoding process and the decoding process.

[Hardware configuration]
FIG. 57 is a diagram showing further details of the smartphone ex115 shown in FIG. FIG. 58 is a diagram illustrating a configuration example of the smartphone ex115. The smartphone ex115 receives an antenna ex450 for transmitting and receiving radio waves to and from the base station ex110, a camera unit ex465 capable of taking video and still images, a video image captured by the camera unit ex465, and an antenna ex450. A display unit ex458 for displaying data obtained by decoding a video or the like. The smartphone ex115 further includes an operation unit ex466 such as a touch panel, a sound output unit ex457 such as a speaker for outputting sound or sound, a sound input unit ex456 such as a microphone for inputting sound, and shooting. Memory unit ex467 that can store encoded data such as encoded video or still images, recorded audio, received video or still images, mail, etc., or decoded data; A slot unit ex464 as an interface unit with the SIMex 468 for authenticating access to various data is provided. Note that an external memory may be used instead of the memory unit ex467.

A main control unit ex460 that can control the display unit ex458, the operation unit ex466, and the like, a power supply circuit unit ex461, an operation input control unit ex462, a video signal processing unit ex455, a camera interface unit ex463, a display control unit ex459; The demodulation unit ex452, the multiplexing / demultiplexing unit ex453, the audio signal processing unit ex454, the slot unit ex464, and the memory unit ex467 are connected via the synchronization bus ex470.

When the power key is turned on by a user operation, the power supply circuit ex461 starts the smartphone ex115 in an operable state, and supplies power from the battery pack to each unit.

The smartphone ex115 performs processing such as telephone communication and data communication based on the control of the main control unit ex460 having a CPU, a ROM, a RAM, and the like. During a call, a voice signal collected by the voice input unit ex456 is converted into a digital voice signal by the voice signal processing unit ex454, a spectrum spread process is performed by the modulation / demodulation unit ex452, and a digital / analog conversion process is performed by the transmission / reception unit ex451. And frequency conversion processing, and transmits the resulting signal via the antenna ex450. Further, the received data is amplified, subjected to frequency conversion processing and analog-to-digital conversion processing, subjected to spectrum despreading processing by a modulation / demodulation unit ex452, converted to an analog audio signal by an audio signal processing unit ex454, and then converted to an audio output unit ex457. Output from In the data communication mode, text, still image, or video data can be transmitted under the control of the main control unit ex460 via the operation input control unit ex462 based on the operation of the operation unit ex466 or the like of the main unit. Similar transmission / reception processing is performed. When transmitting a video, a still image, or a video and audio in the data communication mode, the video signal processing unit ex455 converts the video signal stored in the memory unit ex467 or the video signal input from the camera unit ex465 into each of the above embodiments. The video data is compression-encoded by the moving image encoding method shown in the embodiment, and the encoded video data is transmitted to the multiplexing / demultiplexing unit ex453. The audio signal processing unit ex454 encodes the audio signal collected by the audio input unit ex456 while capturing the video or the still image by the camera unit ex465, and sends out the encoded audio data to the multiplexing / demultiplexing unit ex453. The multiplexing / demultiplexing unit ex453 multiplexes the coded video data and the coded audio data by a predetermined method, and modulates and converts the multiplexed data in the modulation / demodulation unit (modulation / demodulation circuit unit) ex452 and the transmission / reception unit ex451. After processing, the data is transmitted via the antenna ex450. The predetermined method may be determined in advance.

In a case where a video attached to an e-mail or a chat or a video linked to a web page is received, the multiplexing / demultiplexing unit ex453 performs multiplexing / demultiplexing in order to decode the multiplexed data received via the antenna ex450. By separating the multiplexed data, the multiplexed data is divided into a bit stream of video data and a bit stream of audio data, and the coded video data is supplied to the video signal processing unit ex455 via the synchronization bus ex470. The encoded audio data is supplied to the audio signal processing unit ex454. The video signal processing unit ex455 decodes the video signal by the video decoding method corresponding to the video encoding method described in each of the above embodiments, and is linked from the display unit ex458 via the display control unit ex459. The video or still image included in the moving image file is displayed. The audio signal processing unit ex454 decodes the audio signal, and the audio is output from the audio output unit ex457. Due to the increasing popularity of real-time streaming, audio playback may not be socially appropriate in some user situations. Therefore, as an initial value, it is preferable that only the video data is reproduced without reproducing the audio signal, and the audio may be reproduced synchronously only when the user performs an operation such as clicking the video data. .

Although the smartphone ex115 has been described as an example here, as a terminal, in addition to a transmission / reception type terminal having both an encoder and a decoder, a transmission terminal having only an encoder, and a reception having only a decoder are provided. Other implementations, such as terminals, are possible. In the digital broadcasting system, it has been described that multiplexed data in which audio data is multiplexed with video data is received or transmitted. However, the multiplexed data may be multiplexed with character data or the like related to video in addition to audio data. Further, the video data itself may be received or transmitted instead of the multiplexed data.

Although the main control unit ex460 including the CPU controls the encoding or decoding processing, various terminals often include a GPU. Therefore, a configuration in which a wide area is collectively processed by utilizing the performance of the GPU by using a memory shared by the CPU and the GPU or a memory whose addresses are managed so as to be commonly used may be used. As a result, the encoding time can be reduced, real-time performance can be ensured, and low delay can be realized. In particular, it is efficient to perform the motion search, the deblocking filter, the SAO (Sample Adaptive Offset), and the conversion / quantization processing collectively in units of pictures or the like by the GPU instead of the CPU.

The present disclosure is applicable to, for example, a television receiver, a digital video recorder, a car navigation, a mobile phone, a digital camera, a digital video camera, a video conference system, an electronic mirror, and the like.

REFERENCE SIGNS LIST 100 Encoding device 102 Divider 104 Subtractor 106 Transformer 108 Quantizer 110

Entropy encoder

112, 204

Inverse quantizer

114, 206

Inverse transformer

116, 208

Adder

118, 210

Block memory

120, 212

Loop filter Units

122, 214

Frame memory

124, 216

Intra prediction unit

126, 218

Inter prediction unit

128, 220 Prediction control unit 200 Decoding device 202 Entropy decoding unit 1201

Boundary determination unit

1202, 1204, 1206 Switch 1203 Filter determination unit 1205 Filter processing unit 1207 Filter characteristic determining unit 1208 Processing determining unit a1, b1 Processor a2, b2 Memory

Claims

Circuit and
A memory connected to the circuit,
The circuit, in operation,
A plurality of pixel positions that are commonly and relatively determined with respect to both a first range including a first reference block of a current block and a second range including a second reference block of the current block; For each of a plurality of relative pixel positions that are a plurality of pixel positions in each of the range and the second range, the gradient value of the relative pixel position in the first range and the relative pixel position in the second range Deriving the gradient sum absolute value that is the absolute value of the sum of the gradient value and
Deriving, as a first parameter, a sum of a plurality of gradient sum absolute values derived for each of the plurality of relative pixel positions;
For each of the plurality of relative pixel positions, a pixel difference value which is a difference between a pixel value of the relative pixel position in the first range and a pixel value of the relative pixel position in the second range is derived. ,
For each of the plurality of relative pixel positions, a positive / negative sign of a gradient sum that is a sum of a gradient value of the relative pixel position in the first range and a gradient value of the relative pixel position in the second range. Inverting or maintaining the sign of the pixel difference value derived for the relative pixel position,
Each of the plurality of relative pixel positions is derived, and the sum of the plurality of pixel difference values in which the sign is inverted or maintained is derived as a second parameter,
An encoding device that generates a predicted image used for encoding the current block using the first parameter and the second parameter.
For each of the plurality of relative pixel positions,
The gradient value of the relative pixel position in the first range is a horizontal gradient value of the relative pixel position in the first range,
The encoding device according to claim 1, wherein the gradient value of the relative pixel position in the second range is a horizontal gradient value of the relative pixel position in the second range.
The circuit is

Derive the first parameter by

Derive the second parameter by
Ω represents a set of said plurality of relative pixel positions, [i, j] represents each of the plurality of relative pixel positions, for each of the plurality of relative pixel positions, I x 0 is the horizontal gradient value of the relative pixel positions in the first range, I x 1 represents the horizontal gradient value of the relative pixel position in the second range, I 0 is the relative in the first range represents a pixel value of a specific pixel location, I 1 represents the pixel value of the relative pixel position in the second range, abs (I x 1 + I x 0) is the absolute value of I x 1 + I x 0 represents, sign (I x 1 + I x 0) , the encoding apparatus according to claim 2 which represents the sign of I x 1 + I x 0.
For each of the plurality of relative pixel positions,
The gradient value of the relative pixel position in the first range is a vertical gradient value of the relative pixel position in the first range.
The encoding device according to claim 1, wherein the gradient value of the relative pixel position in the second range is a vertical gradient value of the relative pixel position in the second range.
The circuit is

Derive the first parameter by

Derive the second parameter by
Ω represents a set of the plurality of relative pixel positions, [i, j] represents each of the plurality of relative pixel positions, and for each of the plurality of relative pixel positions, I y 0 is: Represents the vertical gradient value of the relative pixel position in the first range, I y 1 represents the vertical gradient value of the relative pixel position in the second range, and I 0 represents the relative gradient value of the relative pixel position in the first range. represents a pixel value of a specific pixel location, I 1 represents the pixel value of the relative pixel position in the second range, abs (I y 1 + I y 0) is the absolute value of I y 1 + I y 0 represents, sign (I y 1 + I y 0) , the encoding apparatus according to claim 4 which represents the sign of I y 1 + I y 0.
The circuit is
For each of the plurality of relative pixel positions, a horizontal value that is the absolute value of the sum of the horizontal gradient value of the relative pixel position in the first range and the horizontal gradient value of the relative pixel position in the second range Deriving a gradient sum absolute value, deriving a sum of a plurality of horizontal gradient sum absolute values respectively derived for the plurality of relative pixel positions,
For each of the plurality of relative pixel positions, the vertical gradient value is the absolute value of the sum of the vertical gradient value of the relative pixel position in the first range and the vertical gradient value of the relative pixel position in the second range. Deriving a gradient sum absolute value, deriving a sum of a plurality of vertical gradient sum absolute values respectively derived for the plurality of relative pixel positions,
When the sum of the absolute values of the plurality of horizontal gradient sums is greater than the sum of the absolute values of the plurality of vertical gradient sums, the gradient value of the relative pixel position in the first range for each of the plurality of relative pixel positions. And applying the horizontal gradient value of the relative pixel position in the first range to the gradient value of the relative pixel position in the second range and the horizontal gradient value of the relative pixel position in the second range. Apply,
When the sum of the plurality of horizontal gradient sum absolute values is not greater than the sum of the plurality of vertical gradient sum absolute values, the gradient of the relative pixel position in the first range for each of the plurality of relative pixel positions. The vertical gradient value of the relative pixel position in the first range is applied to the value, and the vertical gradient value of the relative pixel position in the second range is applied to the gradient value of the relative pixel position in the second range. The encoding device according to claim 1.
The circuit is
Deriving a third parameter using the first parameter as a denominator and using the second parameter as a numerator,
The encoding device according to any one of claims 1 to 6, wherein the prediction image is generated using the third parameter.
The circuit includes, for each of the plurality of relative pixel positions, a gradient value of the relative pixel position in the first range by using a 3-tap filter having a filter coefficient set of [-1, 0, 1]; The encoding device according to any one of claims 1 to 7, wherein a gradient value of the relative pixel position in the second range is derived.
The current block is a sub-block defined by a 4 × 4 pixel size in the encoding unit,
The encoding device according to any one of claims 1 to 8, wherein the circuit derives the first parameter and the second parameter in units of the sub-block.
Circuit and
A memory connected to the circuit,
The circuit, in operation,
A plurality of pixel positions that are commonly and relatively determined with respect to both a first range including a first reference block of a current block and a second range including a second reference block of the current block; For each of a plurality of relative pixel positions that are a plurality of pixel positions in each of the range and the second range, the gradient value of the relative pixel position in the first range and the relative pixel position in the second range Deriving the gradient sum absolute value that is the absolute value of the sum of the gradient value and
Deriving, as a first parameter, a sum of a plurality of gradient sum absolute values derived for each of the plurality of relative pixel positions;
For each of the plurality of relative pixel positions, a pixel difference value that is a difference between a pixel value of the relative pixel position in the first range and a pixel value of the relative pixel position in the second range is derived. ,
For each of the plurality of relative pixel positions, a positive / negative sign of a gradient sum that is a sum of a gradient value of the relative pixel position in the first range and a gradient value of the relative pixel position in the second range. Inverting or maintaining the sign of the pixel difference value derived for the relative pixel position,
Each of the plurality of relative pixel positions is derived, and the sum of a plurality of pixel difference values, each of which is inverted or maintained, is derived as a second parameter,
A decoding device that generates a predicted image used for decoding the current block using the first parameter and the second parameter.
For each of the plurality of relative pixel positions,
The gradient value of the relative pixel position in the first range is a horizontal gradient value of the relative pixel position in the first range,
The decoding device according to claim 10, wherein the gradient value of the relative pixel position in the second range is a horizontal gradient value of the relative pixel position in the second range.
The circuit is

Derive the first parameter by

Derive the second parameter by
Ω represents a set of said plurality of relative pixel positions, [i, j] represents each of the plurality of relative pixel positions, for each of the plurality of relative pixel positions, I x 0 is the horizontal gradient value of the relative pixel positions in the first range, I x 1 represents the horizontal gradient value of the relative pixel position in the second range, I 0 is the relative in the first range represents a pixel value of a specific pixel location, I 1 represents the pixel value of the relative pixel position in the second range, abs (I x 1 + I x 0) is the absolute value of I x 1 + I x 0 represents, sign (I x 1 + I x 0) is the decoding apparatus according to claim 11 which represents the sign of I x 1 + I x 0.
For each of the plurality of relative pixel positions,
The gradient value of the relative pixel position in the first range is a vertical gradient value of the relative pixel position in the first range.
The decoding device according to claim 10, wherein the gradient value of the relative pixel position in the second range is a vertical gradient value of the relative pixel position in the second range.
The circuit is

Derive the first parameter by

Derive the second parameter by
Ω represents a set of the plurality of relative pixel positions, [i, j] represents each of the plurality of relative pixel positions, and for each of the plurality of relative pixel positions, I y 0 is: Represents the vertical gradient value of the relative pixel position in the first range, I y 1 represents the vertical gradient value of the relative pixel position in the second range, and I 0 represents the relative gradient value of the relative pixel position in the first range. represents a pixel value of a specific pixel location, I 1 represents the pixel value of the relative pixel position in the second range, abs (I y 1 + I y 0) is the absolute value of I y 1 + I y 0 represents, sign (I y 1 + I y 0) , the decoding apparatus according to claim 13 which represents the sign of I y 1 + I y 0.
The circuit is
For each of the plurality of relative pixel positions, a horizontal value that is the absolute value of the sum of the horizontal gradient value of the relative pixel position in the first range and the horizontal gradient value of the relative pixel position in the second range Deriving a gradient sum absolute value, deriving a sum of a plurality of horizontal gradient sum absolute values respectively derived for the plurality of relative pixel positions,
For each of the plurality of relative pixel positions, the vertical gradient value is the absolute value of the sum of the vertical gradient value of the relative pixel position in the first range and the vertical gradient value of the relative pixel position in the second range. Deriving a gradient sum absolute value, deriving a sum of a plurality of vertical gradient sum absolute values respectively derived for the plurality of relative pixel positions,
When the sum of the absolute values of the plurality of horizontal gradient sums is greater than the sum of the absolute values of the plurality of vertical gradient sums, the gradient value of the relative pixel position in the first range for each of the plurality of relative pixel positions. And applying the horizontal gradient value of the relative pixel position in the first range to the gradient value of the relative pixel position in the second range and the horizontal gradient value of the relative pixel position in the second range. Apply,
When the sum of the plurality of horizontal gradient sum absolute values is not greater than the sum of the plurality of vertical gradient sum absolute values, the gradient of the relative pixel position in the first range for each of the plurality of relative pixel positions. The vertical gradient value of the relative pixel position in the first range is applied to the value, and the vertical gradient value of the relative pixel position in the second range is applied to the gradient value of the relative pixel position in the second range. The decoding device according to claim 10.
The circuit is
Deriving a third parameter using the first parameter as a denominator and using the second parameter as a numerator,
The decoding device according to any one of claims 10 to 15, wherein the prediction image is generated using the third parameter.
The circuit includes, for each of the plurality of relative pixel positions, a gradient value of the relative pixel position in the first range by using a 3-tap filter having a filter coefficient set of [-1, 0, 1]; The decoding device according to any one of claims 10 to 16, wherein a gradient value of the relative pixel position in the second range is derived.
The current block is a sub-block defined by a 4 × 4 pixel size in the encoding unit,
18. The decoding device according to claim 10, wherein the circuit derives the first parameter and the second parameter in units of the sub-block.
A plurality of pixel positions that are commonly and relatively determined with respect to both a first range including a first reference block of a current block and a second range including a second reference block of the current block; For each of a plurality of relative pixel positions that are a plurality of pixel positions in each of the range and the second range, the gradient value of the relative pixel position in the first range and the relative pixel position in the second range Deriving the gradient sum absolute value that is the absolute value of the sum of the gradient value and
Deriving, as a first parameter, a sum of a plurality of gradient sum absolute values derived for each of the plurality of relative pixel positions;
For each of the plurality of relative pixel positions, a pixel difference value which is a difference between a pixel value of the relative pixel position in the first range and a pixel value of the relative pixel position in the second range is derived. ,
For each of the plurality of relative pixel positions, a positive / negative sign of a gradient sum that is a sum of a gradient value of the relative pixel position in the first range and a gradient value of the relative pixel position in the second range. Inverting or maintaining the sign of the pixel difference value derived for the relative pixel position,
Each of the plurality of relative pixel positions is derived, and the sum of the plurality of pixel difference values in which the sign is inverted or maintained is derived as a second parameter,
An encoding method for generating a predicted image used for encoding the current block using the first parameter and the second parameter.
A plurality of pixel positions that are commonly and relatively determined with respect to both a first range including a first reference block of a current block and a second range including a second reference block of the current block; For each of a plurality of relative pixel positions that are a plurality of pixel positions in each of the range and the second range, the gradient value of the relative pixel position in the first range and the relative pixel position in the second range Deriving the gradient sum absolute value that is the absolute value of the sum of the gradient value and
Deriving, as a first parameter, a sum of a plurality of gradient sum absolute values derived for each of the plurality of relative pixel positions;
For each of the plurality of relative pixel positions, a pixel difference value which is a difference between a pixel value of the relative pixel position in the first range and a pixel value of the relative pixel position in the second range is derived. ,
For each of the plurality of relative pixel positions, a positive / negative sign of a gradient sum that is a sum of a gradient value of the relative pixel position in the first range and a gradient value of the relative pixel position in the second range. Inverting or maintaining the sign of the pixel difference value derived for the relative pixel position,
Each of the plurality of relative pixel positions is derived, and the sum of the plurality of pixel difference values in which the sign is inverted or maintained is derived as a second parameter,
A decoding method for generating a predicted image used for decoding the current block using the first parameter and the second parameter.